CN117711075A - Biological information living body identification method, system, equipment and storage medium - Google Patents

Abstract

A biological information living body recognition method, including: Step S1: Obtain a speckle pattern and detect a target area; the target area is an area containing biological information; Step S2: Identify multiple speckles on the target area and Speckle brightness; Step S3: Calculate the brightness relationship between each speckle and adjacent speckles; Step S4: Determine whether the target object is a living body based on the brightness relationship of multiple speckles and the relationship between the speckle brightness and shape. . The present invention detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles, the relationship between speckle brightness and shape to determine whether the target object is a living body, and has It has the advantages of fast recognition speed, strong adaptability, good biological information recognition effect, and low computing power requirements.

Description

A biological information living identification method, system, equipment and storage medium

Technical field

The present invention relates to the technical field of human biometric identification, and specifically, to a biological information identification method, system, equipment and storage medium.

Background technique

Laser speckle refers to the diffuse reflection of laser light on the surface of a scattering body or when it passes through a transparent scattering body (such as frosted glass). A randomly distributed bright and dark spot can be observed in the light field on or near the scattering surface. This kind of spot It's called laser speckle.

The internal structure of human skin is complex, with capillaries all over the inner surface of the skin. The dermis layer of the skin contains a variety of complex skin tissues (such as nerve fibers, sebaceous glands, hair follicles, sweat glands, etc.), which is a typical anisotropic material. It is currently difficult to make an attack prosthesis with anisotropic properties similar to that of real skin. Attack prostheses currently available on the market are basically made of homogeneous materials. Since the material uniformity of real skin and implants is different in different directions, laser speckle has strong penetrating performance, and the beam will penetrate the stratum corneum and subcutaneous tissue. Factors such as chromophores will affect the scattering and absorption of light, which leads to significant differences in the optical properties of laser speckle absorption and reflection between real skin and attack prostheses.

The liveness determination algorithm is an important step before identifying human biological images, such as face images, palm images, pupil images, fingerprint images, etc. Before identification, the algorithm needs to determine that the image to be recognized belongs to a live real person. It is necessary to resist attacks from printed photos, electronic videos, 3D hand models and other non-real-life images. Currently commonly used palm living algorithms include: RGB-based palm living algorithm, near-infrared based palm living algorithm and depth-based palm living algorithm. For some attack prostheses that are particularly realistic or close to real people (such as 3D silicone gloves, 3D silicone hand models, etc.), the near-infrared or RGB-based in vivo algorithm is prone to misjudgment, while the depth-based in vivo algorithm can withstand 2D prostheses. Attacks, but are often powerless against 3D prosthetic attacks (such as 3D silicone gloves, 3D hand models, etc.). Therefore, the living body determination method based on the speckle pattern proposed by the present invention, as a supplement to the above-mentioned existing methods, captures the essential characteristic difference between the prosthesis attack material and real skin, and has good resistance to various types of different prostheses. ability, which can well make up for the shortcomings of the above-mentioned existing living body determination algorithms.

The disclosure of the above background technology content is only used to assist in understanding the inventive concepts and technical solutions of the present invention. It does not necessarily belong to the prior art of this patent application. In the absence of clear evidence that the above content has been disclosed on the filing date of this patent application, In this case, the above background technology should not be used to evaluate the novelty and inventiveness of this application.

Contents of the invention

To this end, the present invention detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles and the relationship between speckle brightness and shape to determine whether the target object is Living bodies have the advantages of fast recognition speed, strong adaptability, good biological information living body recognition effect, and low computing power requirements.

In a first aspect, the present invention provides a biological information identification method, which is characterized in that it includes:

Step S1: Obtain the speckle pattern and detect the target area; the target area is an area containing biological information;

Step S2: Identify multiple speckles and speckle brightness on the target area;

Step S3: Calculate the brightness relationship between each speckle and adjacent speckles;

Step S4: Determine whether the target object is a living body according to the brightness relationship of multiple speckles and the relationship between the speckle brightness and shape.

Optionally, the biological information identification method is characterized in that step S1 includes:

Step M11: Collect speckle images;

Step M12: Preprocess the collected speckle images;

Step M13: Extract features from the preprocessed speckle image;

Step M14: Match the extracted features with the known target area;

Step M15: Output the matching results.

Optionally, the biological information identification method is characterized in that step S2 includes:

Step S21: Detect edges in the speckle pattern according to an edge detection algorithm and obtain the position of each speckle;

Step S22: Perform expansion and erosion operations on the edge to obtain the precise position of each speckle;

Step S23: Mark the position of each speckle to determine its specific position in the image;

Step S24: Calculate the average brightness value and brightness distribution proportion of each speckle based on the specific location of each speckle.

Optionally, the biological information identification method is characterized in that step S3 includes:

Step S31: Find multiple adjacent speckles according to the position of each speckle and under distance constraints;

Step S32: Multiply the difference vector of the maximum brightness of the speckle and the adjacent speckle to obtain the high difference vector of the speckle;

Step S33: Multiply the difference vector of the average brightness of the speckle and the adjacent speckle to obtain the mean difference vector of the speckle.

Optionally, the biological information living body identification method is characterized in that in step S4, a deep learning model is used to determine whether the target object is a living body. The neural network of the deep learning model during training includes: a first inverse reaction. Directional residual module, first pooling layer, second reverse residual module, second pooling layer, third reverse residual module, first overlay layer, third pooling layer, fourth reverse residual module Module, second stacking layer, fourth pooling layer, fifth inverse residual module, third convolutional layer, fifth pooling layer, random dropout layer, classifier, first convolutional layer, second convolution layer; wherein, the first reverse residual module outputs to the first pooling layer and the first convolution layer, and the first convolution layer and the third reverse residual module input the The first overlay layer, the second reverse residual module output to the second convolution layer and the second pooling layer, the second convolution layer and the fourth reverse residual module The second overlay is input; the classifier outputs a live probability and a prosthetic probability.

Optionally, the biological information identification method is characterized in that step S4 includes:

Step T41: Count the brightness relationships of multiple speckles. If the preset conditions are met, step T42 is executed. Otherwise, it is determined to be a prosthesis;

Step T42: Calculate the relationship between the average brightness of the speckle and the shape. If it meets the preset conditions, execute step T43; otherwise, it is determined to be a prosthesis;

Step T43: Calculate the relationship between the brightness distribution proportion of the speckle and the shape. If it meets the preset conditions, it is determined to be a living body, otherwise it is determined to be a prosthesis.

Optionally, the biological information identification method is characterized in that step S4 includes:

Step S41: Calculate the relationship between the speckle brightness and shape. If it meets the preset conditions, execute step S42; otherwise, it is determined to be a prosthesis;

Step S42: Statistics the distribution of the high difference vectors to obtain the number m of speckles greater than the first threshold;

Step S43: If m is greater than the second threshold, execute step S44, otherwise it is determined to be a prosthesis;

Step S44: Statistics the distribution of the mean difference vector, and obtains the number n of speckles greater than the third threshold;

Step S45: If n is greater than the fourth threshold, it is determined to be a living body, otherwise it is determined to be a prosthesis.

In a second aspect, the present invention provides a biological information identification system for implementing any of the aforementioned biological information identification methods, which is characterized in that it includes:

An acquisition module is used to acquire a speckle pattern and detect a target area; the target area is an area containing biological information;

An identification module, used to identify multiple speckles and speckle brightness on the target area;

A calculation module used to calculate the brightness relationship between each speckle and adjacent speckles;

A judgment module configured to judge whether the target object is a living body according to the brightness relationship of multiple speckles and the relationship between the speckle brightness and shape.

In a third aspect, the present invention provides a biological information identification device, which is characterized in that it includes:

processor;

A memory in which executable instructions of the processor are stored;

Wherein, the processor is configured to execute any of the foregoing steps of the biological information identification method by executing the executable instructions.

In a fourth aspect, the present invention provides a computer-readable storage medium for storing a program, characterized in that when the program is executed, the steps of any one of the aforementioned biological information identification methods are implemented.

Compared with the prior art, the present invention has the following beneficial effects:

The present invention uses speckle patterns to directly conduct living body detection, taking advantage of the different effects of speckles at different depths and surfaces of different materials. Based on the brightness relationship and shape between multiple speckles, it is judged whether the target object is a living body, which is better than using a single speckle pattern. Speckle characteristics have better results in living body recognition. At the same time, because of the comparison between multiple speckles, it can also better identify attacks on three-dimensional prostheses.

The invention can make prosthetic judgments on three-dimensional silicone hand models, three-dimensional plastic hand models, three-dimensional rubber hand models, three-dimensional silicone sleeves, and two-dimensional planar prosthetic hands, and can effectively grasp the essential characteristics of living real hands and attack the prostheses. It has a wide range of resistance to various types of attack prostheses.

The invention can effectively reduce the impact of ambient lighting on the algorithm, and can work stably without being affected during the day and night and under various extreme lighting conditions.

Description of the drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without exerting creative efforts. Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

Figure 1 is a step flow chart of a biological information identification method in an embodiment of the present invention;

Figure 2 is a speckle pattern in an embodiment of the present invention;

Figure 3 is a flow chart of steps for detecting a target area in an embodiment of the present invention;

Figure 4 is a flow chart of steps for identifying speckles and speckle brightness in an embodiment of the present invention;

Figure 5 is a flow chart of steps for calculating the brightness relationship between speckles and adjacent speckles in an embodiment of the present invention;

Figure 6 is a schematic structural diagram of a neural network in an embodiment of the present invention;

Figure 7 is a schematic structural diagram of a reverse residual module in an embodiment of the present invention;

Figure 8 is a flow chart of steps for determining whether it is a living body in an embodiment of the present invention;

Figure 9 is another step flow chart for determining whether it is a living body in an embodiment of the present invention;

Figure 10 is a schematic structural diagram of a biological information identification system in an embodiment of the present invention;

Figure 11 is a schematic structural diagram of a biological information identification device in an embodiment of the present invention; and

Figure 12 is a schematic structural diagram of a computer-readable storage medium in an embodiment of the present invention.

Detailed ways

The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

The terms "first", "second", "third", "fourth", etc. (if present) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects without necessarily using Used to describe a specific order or sequence. It is to be understood that the figures so used are interchangeable under appropriate circumstances so that the embodiments of the invention described herein, for example, can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "include" and "having" and any variations thereof are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus that encompasses a series of steps or units and need not be limited to those explicitly listed. Those steps or elements may instead include other steps or elements not expressly listed or inherent to the process, method, product or apparatus.

The embodiment of the present invention provides a biological information living body recognition depth module method, aiming to solve the problems existing in the existing technology.

The technical solution of the present invention and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will be described below with reference to the accompanying drawings.

The present invention detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles, the relationship between speckle brightness and shape to determine whether the target object is a living body, and has It has the advantages of fast recognition speed, strong adaptability, good biological information recognition effect, and low computing power requirements.

Figure 1 is a step flow chart of a biological information identification method in an embodiment of the present invention. As shown in Figure 1, the steps of a biological information identification method in an embodiment of the present invention include:

Step S1: Obtain the speckle image and detect the target area.

In this step, the target area is an area containing biological information. A speckle image is an image containing multiple speckles. Speckles can be sparse or dense. When detecting the target area, the speckle pattern can be used directly for detection, or other images can be used for detection, and then the target area can be obtained based on the corresponding relationship between the images. For example, when using other images for detection, the speckle image and the infrared image are obtained at the same time, and the target detection model is used on the infrared image to obtain the target area on the infrared image, and then the speckle image is obtained based on the correspondence between the speckle image and the infrared image. target area. In order to facilitate calculation, an infrared collector can be used to obtain speckle and flood signals at the same time. The generated image contains both speckle information and infrared information. By extracting the generated image, the speckle map and Infrared image, so that the infrared image and the speckle image achieve a one-to-one correspondence at the pixel level.

Figure 2 shows the speckle imaging of single-point laser speckle spots on real skin and three attack prostheses. In Figure 2, all the shooting samples are at the same distance from the laser and camera, and the imaging exposure time is the same. It can be clearly seen that the intensity distribution of laser speckles on different materials is obviously different, and the speckles on different materials are clear. There is also a relatively big difference in degree.

In some embodiments, this step includes:

Step M11: Collect speckle images. These images should contain the target area. The acquisition method can be optical imaging, radar scanning, etc.

Step M12: Preprocess the collected speckle images. Preprocessing methods include denoising, smoothing, enhancement and other operations to improve the accuracy and efficiency of subsequent processing.

Step M13: Extract features from the preprocessed speckle image. Commonly used features include edge features, corner features, texture features, etc. These features are used to distinguish different target areas.

Step M14: Match the extracted features with the known target area. Template matching-based methods or deep learning-based methods can be used. If the match is successful, the target area has been detected.

Step M15: Output the matching results. The output result can be a binary image or a target area represented by a polygon bounding box.

Step S2: Identify multiple speckles and speckle brightness on the target area.

In this step, there are multiple speckles in the target area. As shown in Figure 2, there are obvious differences between the speckles irradiated on the real person and the prosthesis. At the same time, because human body features such as faces and palms are in three-dimensional space, their surfaces are not flat, but have different slopes and orientations. In addition, there are differences in the skin in different areas, making each speckle have its own characteristics. The characteristic, specifically reflected in the image, is that the shape and brightness of the speckles are different. Form includes shape and size. Brightness includes the overall brightness and the attenuation rate of brightness from the center to the surroundings.

Step S3: Calculate the brightness relationship between each speckle and adjacent speckles.

In this step, since there are multiple speckles around each speckle, it is necessary to find the adjacent relationship between the speckle and the surrounding speckles. When determining the adjacent speckles of a speckle, the final determination is based on the design of the speckle. For example, when the speckle pattern is a square pattern, the center of the speckle is the origin, and 90 degrees is the interval to find the nearest speckle, so that in any 90-degree direction, there is a speckle as an adjacent speckle.

According to the brightness relationship between the speckle and adjacent speckles, the brightness relationship matrix of the speckle can be obtained. For example, if the brightness of a speckle is 100, and the brightness of adjacent speckles are 80, 102, 89, and 120 respectively, then the brightness relationship matrix is [-20, 2, -11, 20]. The brightness relationship matrix can characterize changes in brightness and be used to identify prostheses.

Step S4: Determine whether the target object is a living body according to the brightness relationship of multiple speckles and the relationship between the speckle brightness and shape.

In this step, based on the relationship between the shape and brightness of all speckles in the speckle diagram and the brightness relationship matrix, it can be judged whether it conforms to the characteristics of a living body, and thereby whether it is a living body. When judging whether it is a living body, either the threshold method or the deep learning model method can be used. When using the threshold method, the number of speckles that match the characteristics of a real person is counted, and when the number is more than the threshold, it is determined to be a living body. When a deep learning model is used, the features obtained in steps S2 and S3 are input into the deep learning model for training, and the final model is obtained for judgment. The relationship between the speckle brightness and shape shows a specific relationship. Usually in a depth camera, multiple speckles projected by a speckle projector have the same shape and projection intensity. However, due to different characteristics of the target object, such as dermal tissue, bones under the skin, etc., have different effects on the speckles, resulting in In the speckle pattern of the human body, there will be obvious changes in the brightness and shape of different speckles. However, the material of the prosthesis is relatively consistent and does not have the above characteristics.

Figure 3 is a flow chart of steps for detecting a target area in an embodiment of the present invention. As shown in Figure 3, a step of detecting a target area in an embodiment of the present invention includes:

Step S11: Obtain the speckle pattern.

In this step, a depth camera can be used to obtain the speckle image, such as a structured light camera.

Step S12: Detect the speckle pattern according to the target detection model to obtain the first target area.

In this step, the target detection model is used to process the speckle image to identify the target object in it. The target detection model is a model based on machine learning algorithms that can automatically identify the target object based on the input image data and give information such as its location and size. In this step, the already trained target detection model can be used to detect the speckle pattern to obtain the first target area.

Step S13: Intercept and obtain the second target area based on the key points on the first target area.

In this step, according to the key points on the first target area, a small part of the area needs to be intercepted as the second target area. Key points refer to points with special significance in the first target area. The selection of key points is based on the type of target object. For example, 86 key points can be selected for the face, 7 or 21 key points can be used for the palm, etc. By intercepting the area around these key points, the location and size of the target object can be more accurately located. It should be noted that the second target area is a continuous and complete area, including the effective area of exposed skin in other areas such as the face or palms.

This step obtains a smaller second target area based on the key points, focusing on the exposed skin area, and filtering out hair, decoration and other areas, so that the speckles in the second target area are obtained after reflection from the skin, so that The proportion of data for subsequent processing is guaranteed.

Figure 4 is a flow chart of steps for identifying speckles and speckle brightness in an embodiment of the present invention. As shown in Figure 4, a step of identifying speckles and speckle brightness in an embodiment of the present invention includes:

Step S21: Detect edges in the speckle image according to an edge detection algorithm to obtain the position of each speckle.

In this step, you need to use an edge detection algorithm to process the speckle image to identify the edges. The edge detection algorithm is an algorithm based on image processing technology that can automatically identify edge information in images. Common edge detection algorithms (such as Sobel operator, Canny operator, etc.) can be used to process the speckle image to obtain the position of each speckle.

Step S22: Perform expansion and erosion operations on the edge to obtain the precise position of each speckle.

In this step, the edges need to be expanded and eroded to obtain the precise location of each speckle. The dilation operation can expand the edge outward by a certain pixel value to obtain more accurate boundary information; the erosion operation can shrink the edge inward by a certain pixel value to remove some noise information. Commonly used expansion and erosion operation methods (such as opening operations, closing operations, etc.) can be used to process the edges to obtain the precise position of each speckle.

Step S23: Position mark each speckle to determine its specific position in the image.

In this step, each speckle needs to be positioned to determine its specific location in the image. The coordinate system in computer vision technology can be used to mark the position of each speckle. For example, the upper left corner of the image is used as the origin, the positive x-axis direction is to the right, and the positive y-axis direction is downward. Programming languages (such as Python, MATLAB, etc.) can be used to implement the position marking function.

Step S24: Calculate the average brightness value and brightness distribution proportion of each speckle based on the specific location of each speckle.

In this step, it is necessary to calculate the average brightness value and brightness distribution proportion of each speckle based on its specific location. The average brightness value refers to the average brightness of all pixels in the speckle area; the brightness distribution proportion refers to the proportion of different brightness values in the speckle area. Programming languages (such as Python, MATLAB, etc.) can be used to realize the calculation function of average brightness value and brightness distribution proportion.

This embodiment uses an edge detection algorithm to effectively identify the location of the target object from the speckle pattern, thereby providing a basis for subsequent processing; by performing expansion and erosion operations on the edges, the precise location of each speckle can be obtained. Improve the accuracy of positioning; by marking the position of each speckle, we can help us determine the specific location of the target object in the image, which is crucial for the tracking and analysis of the target object; according to the specific location of each speckle, Calculating the average brightness value and brightness distribution proportion of each speckle can better evaluate the speckle pattern.

Figure 5 is a flow chart of steps for calculating the brightness relationship between speckles and adjacent speckles in an embodiment of the present invention. As shown in Figure 5, a step of calculating the brightness relationship between a speckle and an adjacent speckle in an embodiment of the present invention includes:

Step S31: Find multiple adjacent speckles according to the position of each speckle and under distance constraints.

In this step, the distance constraint means that the distance between adjacent speckles must be within a certain range, otherwise they will be regarded as non-adjacent. A certain range depends on the speckle design. For example, within the effective depth range, the design spacing of speckles is 2mm, then take the center of the speckles as the center of the circle, search within a circle with a radius of 4mm (twice the design spacing), and obtain all speckles, which is the phase Orbital speckle. Programming languages (such as Python, MATLAB, etc.) can be used to implement the search function of adjacent speckles.

Step S32: Multiply the difference vector of the maximum brightness of the speckle and the adjacent speckle to obtain a high difference vector of the speckle.

In this step, the high difference vector refers to the degree of difference between different brightness values in the speckle area. This step is calculated based on the maximum brightness of the speckle and the maximum brightness of adjacent speckles, which can well represent the changes in the maximum brightness of different speckles. In this step, you can use programming languages (such as Python, MATLAB, etc.) to implement the calculation function of high difference vectors.

Step S33: Multiply the difference vector of the average brightness of the speckle and the adjacent speckle to obtain the mean difference vector of the speckle.

In this step, the average difference vector refers to the degree of difference between the average brightness of all pixels in the speckle area and the average brightness of adjacent pixels. This step is calculated based on the average brightness of the speckle and the average brightness of adjacent speckles, which can well represent the changes in the average brightness of different speckles. In this step, you can use a programming language (such as Python, MATLAB, etc.) to implement the calculation function of the mean difference vector.

This embodiment calculates a high difference vector and a mean difference vector, which can better characterize the brightness relationship between speckles and adjacent speckles, thereby improving the accuracy of evaluation.

Figure 6 is a schematic structural diagram of a neural network in an embodiment of the present invention. As shown in Figure 6, a neural network in the embodiment of the present invention is used to train a deep learning model, including: a first reverse residual module, a first pooling layer, a second reverse residual module, and a second pooling layer. layer, the third reverse residual module, the first overlay layer, the third pooling layer, the fourth reverse residual module, the second overlay layer, the fourth pooling layer, the fifth reverse residual module, the third Convolutional layer, fifth pooling layer, random discarding layer, classifier, first convolutional layer, second convolutional layer; wherein the first reverse residual module outputs to the first pooling layer and The first convolution layer, the first convolution layer and the third inverse residual module input the first overlay layer, and the second inverse residual module outputs to the second convolution layer and the second pooling layer, the second convolution layer and the fourth inverse residual module input the second overlay layer; the classifier outputs the probability of living body and the probability of prosthesis.

It should be noted that those skilled in the art can adjust the number of pooling layers according to the image size, thereby striking a balance between training speed and training effect.

The neural network design during training of deep learning models has the following significant advantages:

Layer-by-layer pre-training: The deep learning model adopts a layer-by-layer pre-training training mechanism, which enables it to overcome the problems of easy overfitting and slow training speed of traditional neural networks.

Automatic feature learning: Through layer-by-layer data pre-training, the deep learning model can automatically learn the primary features of the data, thus overcoming the time-consuming and laborious traditional method of manually designing features.

Distributed data learning: Deep learning models can learn more effectively on distributed data, making learning efficiency exponential.

Deep modeling capabilities: Compared with shallow modeling methods, deep modeling can represent actual complex nonlinear problems in more detail and efficiently. This means that deep learning models have stronger performance when dealing with complex real-world problems, such as image recognition, speech recognition, etc.

This embodiment enables the deep learning model to have higher efficiency and accuracy when processing complex tasks.

Figure 7 is a schematic structural diagram of a reverse residual module in an embodiment of the present invention. As shown in Figure 7, a reverse residual module in the embodiment of the present invention includes: a first sub-convolution layer, a first batch normalization layer, a first non-linear activation layer, a second sub-convolution layer, two batch normalization layers, a second nonlinear activation layer, a third sub-convolution layer, a third batch normalization layer, and a first sub-overlay layer; wherein the first sub-convolution layer is output to the A batch normalization layer, the first sub-convolutional layer and the third batch normalization layer are input to the first overlay layer.

The reverse residual module in this embodiment may be the first reverse residual module, the second reverse residual module, the third reverse residual module, the fourth reverse residual module in the previous embodiment. Any one or more of the residual module and the fifth reverse residual module.

This embodiment has at least the following obvious advantages:

Reduce the number of model parameters: The inverse residual module can effectively reduce the number of model parameters by performing a reversible transformation on the residual connection. Traditional residual modules require a large number of convolutional layers, which require a large number of parameters. The inverse residual module effectively reduces the number of parameters by inverting the residual structure, that is, first performing projected convolution to increase the dimension, then using depth convolution, and finally using projected convolution to reduce the dimension.

Improve model performance: The design of the inverse residual module prevents the network from disappearing or exploding gradients during training, allowing the training of very deep networks. This helps improve model performance.

Low implementation complexity: The implementation of the inverse residual module is mainly completed through two layers of the same or different convolutional layers and a skip connection. Its implementation is relatively simple and the computational complexity is low.

Therefore, the embodiment effectively solves the gradient problem and the problem of too many parameters in deep network training, improves the performance of the model, and is relatively simple to implement, which makes it of high practical value when processing complex tasks.

Figure 8 is a flow chart of steps for determining whether a living body is alive in an embodiment of the present invention. As shown in Figure 8, a step of determining whether it is a living body in the embodiment of the present invention includes:

Step T41: Count the brightness relationships of multiple speckles. If the preset conditions are met, step T42 is executed. Otherwise, it is determined to be a prosthesis.

Specifically, according to the different characteristics of the human body, such as the face, palms, pupils, etc., the brightness distribution of speckles at different locations will be different. According to the position of the human body represented by the speckles and the ordering of the brightness among all speckles , can determine whether it is a living body. For example, the speckle brightness at the human face position is 150, and the brightness is in the top 1/3 of all speckles, which meets the preset conditions of (1/4, 2/5) for this position. All speckle spots are judged. If the proportion of qualified speckle spots in all speckle spots reaches a preset ratio, such as more than 90%, proceed to the next step, otherwise it is judged to be a prosthesis.

Step T42: Statistics of the relationship between the average brightness of the speckle and the shape are performed. If the preset conditions are met, step T43 is executed. Otherwise, it is determined to be a prosthesis.

In this step, the average brightness refers to the average value of multiple speckle brightness values. Using the shape as a reference, obtain the average value of the brightness of multiple speckles in the same shape, thereby obtaining a curve between shape and average brightness. If the curve meets the expected conditions, step M42 is executed, otherwise it is determined to be a prosthesis. Shape can use quantitative indicators directly related to the shape, such as the number of pixels, speckle radius, etc.

Step T43: Calculate the relationship between the brightness distribution proportion of the speckle and the shape. If it meets the preset conditions, it is determined to be a living body, otherwise it is determined to be a prosthesis.

In this step, similar to step M42, the corresponding relationship between the brightness distribution proportion and the shape is calculated. Using the shape as a reference, obtain the number of pixels with different brightness in multiple speckles under the same depth value, thereby obtaining the brightness distribution proportion corresponding to the shape. Compare the brightness distribution proportions of different shapes with the preset conditions. If they match, it is judged to be a living body, otherwise it is judged to be a prosthesis.

This embodiment determines based on the brightness relationship, shape and average brightness, and brightness distribution proportion, and obtains the impact of depth value on brightness, so that the impact of brightness can be evaluated at different depths, and the impact of different materials can be identified, which has a very high accuracy. Recognition accuracy.

Figure 9 is another flow chart of steps for determining whether a living body is alive in an embodiment of the present invention. As shown in Figure 9, another step of determining whether it is a living body in the embodiment of the present invention includes:

Step S41: Calculate the relationship between the speckle brightness and shape. If it meets the preset conditions, execute step S42; otherwise, it is determined to be a prosthesis.

In this step, when the speckle is irradiated on different parts of the human body, it will have different effects on the light spot due to the influence of the surface tissue of the human body, resulting in different shapes and spot brightness. In this step, the relationship between the shape of the speckle and the brightness is calculated, and a relationship curve is obtained, so that judgment can be made based on the relationship curve. If the relationship curve meets the preset conditions, for example, compared with the preset curve, the error is within the specified range, then proceed to the next step, otherwise, it is determined to be a prosthesis.

Step S42: Statistics the distribution of the high difference vectors to obtain the number m of speckles greater than the first threshold.

In this step, statistical analysis is first performed on the high difference vector, and the number m of speckles greater than the first threshold is calculated. The distribution of high difference vectors can be used to directly filter out prostheses that have relatively consistent surface properties but are quite different from the human body.

Step S43: If m is greater than the second threshold, execute step S44, otherwise it is determined to be a prosthesis.

In this step, if the number m of speckles greater than the first threshold is greater than the second threshold, it indicates that there may be a problem with the model and further inspection is required. If m is not greater than the second threshold, then we can initially determine that the model is a prosthesis.

Step S44: Statistics the distribution of the mean difference value vector, and obtain the number n of speckles greater than the third threshold.

In this step, continue to perform statistical analysis on the mean difference vector, and calculate the number n of speckles greater than the third threshold. The distribution of the mean difference value vector is used to further filter out prostheses with inconsistent surface properties.

Step S45: If n is greater than the fourth threshold, it is determined to be a living body, otherwise it is determined to be a prosthesis.

In this step, the judgment is made based on the distribution of the mean difference vector. If the number n of speckles greater than the third threshold is greater than the fourth threshold, we consider the model to be living. If it is not greater than the fourth threshold, the model is determined to be a prosthesis.

This embodiment uses high difference vectors and mean difference vectors to filter out prostheses whose surface properties are relatively consistent but are quite different from the human body, and prostheses whose surface properties are inconsistent, and the surface material distribution is judged. This can resist various types of prosthetic attacks, greatly improve the identification effect of prosthetic attacks, and has good identification of various types of prostheses, which can greatly improve security.

Figure 10 is a schematic structural diagram of a biological information identification system in an embodiment of the present invention. As shown in Figure 10, a biological information identification system in an embodiment of the present invention includes:

An acquisition module is used to acquire a speckle pattern and detect a target area; the target area is an area containing biological information;

An identification module, used to identify multiple speckles and speckle brightness on the target area;

A calculation module used to calculate the brightness relationship between each speckle and adjacent speckles;

A judgment module configured to judge whether the target object is a living body according to the brightness relationship of multiple speckles and the relationship between the speckle brightness and shape.

This embodiment detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles and the relationship between speckle brightness and shape to determine whether the target object is a living body. It has the advantages of fast recognition speed, strong adaptability, good biological information recognition effect, and low computing power requirements.

An embodiment of the present invention also provides a biological information identification device, including a processor. Memory, which stores the executable instructions of the processor. Wherein, the processor is configured to execute the steps of a biological information identification method by executing executable instructions.

As above, this embodiment detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles and the relationship between speckle brightness and shape to determine whether the target object is Living bodies have the advantages of fast recognition speed, strong adaptability, good biological information living body recognition effect, and low computing power requirements.

Those skilled in the art will understand that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention can be implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "Circuit", "Module" or "Platform".

Figure 11 is a schematic structural diagram of a biological information identification device in an embodiment of the present invention. An electronic device 600 according to this embodiment of the present invention is described below with reference to FIG. 11 . The electronic device 600 shown in FIG. 11 is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

As shown in Figure 11, electronic device 600 is embodied in the form of a general computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, and the like.

The storage unit stores program code, and the program code can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above-mentioned biological information identification method part of this specification. For example, processing unit 610 may perform steps as shown in FIG. 1 .

The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and/or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

Storage unit 620 may also include a program/utility 6204 having a set of (at least one) program modules 6205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples, or some combination, may include the implementation of a network environment.

Bus 630 may be a local area representing one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or using any of a variety of bus structures. bus.

Electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with electronic device 600, and/or with Any device (eg, router, modem, etc.) that enables the electronic device 600 to communicate with one or more other computing devices. This communication may occur through input/output (I/O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 660. Network adapter 660 may communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in Figure 11, other hardware and/or software modules may be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tapes Drives and data backup storage platforms, etc.

Embodiments of the present invention also provide a computer-readable storage medium for storing a program, and the steps of a biological information identification method are implemented when the program is executed. In some possible implementations, various aspects of the present invention can also be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to cause the terminal device to execute the above described instructions. The steps according to various exemplary embodiments of the present invention are described in the section "A Biometric Information Life Recognition Method".

As shown above, this embodiment detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles and the relationship between speckle brightness and shape to determine the target object. Whether it is a living body or not, it has the advantages of fast recognition speed, strong adaptability, good biological information living body recognition effect, and low computing power requirements.

Figure 12 is a schematic structural diagram of a computer-readable storage medium in an embodiment of the present invention. Referring to FIG. 12 , a program product 800 for implementing the above method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be used on a terminal device, For example, run on a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus or device.

The Program Product may take the form of one or more readable media in any combination. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: electrical connection with one or more conductors, portable disk, hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.

A computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave carrying the readable program code therein. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on a readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., as well as conventional procedural programming. Language—such as "C" or a similar programming language. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server execute on. In situations involving remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., provided by an Internet service). (business comes via Internet connection).

This embodiment detects the target area on the speckle map, obtains the brightness relationship between multiple speckles and adjacent speckles, and then uses the brightness relationship between multiple speckles and the relationship between speckle brightness and shape to determine whether the target object is a living body. It has the advantages of fast recognition speed, strong adaptability, good biological information recognition effect, and low computing power requirements.

Each embodiment in this specification is described in a progressive manner. Each embodiment focuses on its differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. The above description of the disclosed embodiments enables those skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be practiced in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above. Those skilled in the art can make various variations or modifications within the scope of the claims, which does not affect the essence of the present invention.

Claims (10)

1. A biological information living body identification method, characterized by comprising:

step S1: acquiring a speckle pattern, and detecting a target area; the target area is an area containing biological information;

step S2: identifying a plurality of speckle and speckle intensities on the target region;

step S3: calculating the brightness relation between each speckle and adjacent speckle;

step S4: and judging whether the target object is a living body according to the brightness relation of the speckles and the relation of the speckles brightness and the shape.

2. The method for identifying a living organism according to claim 1, wherein the step S1 comprises:

step M11, collecting speckle images;

step M12, preprocessing the collected speckle image;

m13, extracting features from the preprocessed speckle images;

m14, matching the extracted features with known target areas;

and M15, outputting a matching result.

3. The method for identifying a living organism according to claim 1, wherein the step S2 comprises:

step S21: detecting edges in the speckle pattern according to an edge detection algorithm to obtain the position of each speckle;

step S22: performing expansion and corrosion operation on the edge to obtain the accurate position of each speckle;

Step S23: position marking is carried out on each speckle, and the specific position in the image is determined;

step S24: and calculating the average brightness value and the brightness distribution proportion of each speckle according to the specific position of each speckle.

4. The method for identifying a living organism according to claim 1, wherein the step S3 comprises:

step S31: according to the position of each speckle, a plurality of adjacent speckles are found under the distance constraint;

step S32: multiplying the speckle with the difference vector of the maximum brightness of the adjacent speckle to obtain a high difference vector of the speckle;

step S33: and multiplying the speckle by the difference vector of the average brightness of the adjacent speckle to obtain the average difference vector of the speckle.

5. The method according to claim 1, wherein in step S4, a deep learning model is used to determine whether the target object is a living body, and the neural network of the deep learning model during training includes: the device comprises a first reverse residual error module, a first pooling layer, a second reverse residual error module, a second pooling layer, a third reverse residual error module, a first superposition layer, a third pooling layer, a fourth reverse residual error module, a second superposition layer, a fourth pooling layer, a fifth reverse residual error module, a third convolution layer, a fifth pooling layer, a random discarding layer, a classifier, a first convolution layer and a second convolution layer; the first reverse residual error module is output to the first pooling layer and the first convolution layer, the first convolution layer and the third reverse residual error module input the first superposition layer, the second reverse residual error module is output to the second convolution layer and the second pooling layer, and the second convolution layer and the fourth reverse residual error module input the second superposition layer; the classifier outputs a living probability and a prosthesis probability.

6. A method of biometric identification according to claim 3, wherein step S4 comprises:

step T41: counting the brightness relation of a plurality of speckles, executing a step T42 if the brightness relation accords with a preset condition, otherwise, judging the brightness relation as a prosthesis;

step T42: counting the relation between the average brightness and the shape of the speckles, if the relation meets the preset condition, executing a step T43, otherwise, judging the relation as a prosthesis;

step T43: and counting the relation between the brightness distribution proportion and the shape of the speckles, judging as a living body if the relation meets preset conditions, and judging as a prosthesis if the relation is not met.

7. The method for identifying a living organism according to claim 4, wherein step S4 comprises:

step S41: counting the relation between the brightness and the shape of the speckles, executing the step S42 if the relation meets the preset condition, otherwise, judging the relation as a prosthesis;

step S42: counting the distribution of the Gao Chazhi vectors to obtain the speckle quantity m larger than a first threshold value;

step S43: if m is greater than the second threshold, executing step S44, otherwise, judging as a prosthesis;

step S44: counting the distribution of the mean difference value vector to obtain the speckle number n larger than a third threshold value;

step S45: if n is greater than the fourth threshold, a living body is determined, otherwise a prosthesis is determined.

8. A bioinformatic living body identification system for realizing the bioinformatic living body identification method according to any one of claims 1 to 7, characterized by comprising:

the acquisition module is used for acquiring the speckle pattern and detecting a target area; the target area is an area containing biological information;

the identification module is used for identifying a plurality of speckles and speckle brightness on the target area;

the calculating module is used for calculating the brightness relation between each speckle and the adjacent speckle;

and the judging module is used for judging whether the target object is a living body according to the brightness relation of the plurality of speckles and the relation of the brightness and the shape of the speckles.

9. A bioinformatic living body identification apparatus, characterized by comprising:

a processor;

a memory having stored therein executable instructions of the processor;

wherein the processor is configured to perform the steps of the method for bioinformatic in-vivo identification of any one of claims 1 to 7 via execution of the executable instructions.

10. A computer-readable storage medium storing a program, characterized in that the program when executed implements the steps of the biological information living body identification method according to any one of claims 1 to 7.