WO2017028742A1 - 图像去噪系统及图像去噪方法 - Google Patents

图像去噪系统及图像去噪方法 Download PDF

Info

Publication number
WO2017028742A1
WO2017028742A1 PCT/CN2016/094817 CN2016094817W WO2017028742A1 WO 2017028742 A1 WO2017028742 A1 WO 2017028742A1 CN 2016094817 W CN2016094817 W CN 2016094817W WO 2017028742 A1 WO2017028742 A1 WO 2017028742A1
Authority
WO
WIPO (PCT)
Prior art keywords
current pixel
pixel point
sobel operator
denoising
sobel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2016/094817
Other languages
English (en)
French (fr)
Inventor
毛水江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BYD Co Ltd
Original Assignee
BYD Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by BYD Co Ltd filed Critical BYD Co Ltd
Priority to KR1020187003980A priority Critical patent/KR20180033211A/ko
Priority to US15/748,699 priority patent/US20190005624A1/en
Priority to EP16836606.0A priority patent/EP3340168A4/en
Publication of WO2017028742A1 publication Critical patent/WO2017028742A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/47—Image sensors with pixel address output; Event-driven image sensors; Selection of pixels to be read out based on image data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise

Definitions

  • the present application relates to the field of image processing, and more particularly to an image denoising system and an image denoising method.
  • existing image denoising methods include local adaptive denoising, variance based denoising, wavelet transform denoising, median filtering denoising, and the like.
  • denoising because the noise is in the high frequency region, it is easy to be confused with the image details, and the loss of detail is caused by denoising, which makes the image blurred;
  • some denoising algorithms such as Wavelet edge denoising
  • CMOS and CCD which have been designed in low cost in recent years.
  • the present application is intended to address at least one of the technical problems existing in the prior art.
  • the present application provides an image denoising system and an image denoising method.
  • An image denoising system comprising: an image signal acquiring unit, wherein the image signal acquiring unit is configured to acquire and output an image signal; and a processing unit, wherein the processing unit is configured to receive the image signal and perform the image signal Pre-processing to obtain image data, and performing SOBEL operator calculation on the current pixel point in the image data to obtain a SOBEL operator of the current pixel point, and correcting the SOBEL operator of the current pixel point to obtain a correction a subsequent SOBEL operator, and determining whether the current pixel point needs to be denoised according to the corrected SOBEL operator; the processing unit is further configured to: when determining that the current pixel point needs to perform denoising processing, Calculating an average pixel value of the pixel points around the current pixel point, and performing denoising processing on the current pixel point by using the average pixel value and the corrected SOBEL operator; the processing unit is further configured to Judging that the current pixel point does not need to perform denoising processing Whether the next pixel in the image data
  • An image denoising method includes: acquiring and outputting an image signal; receiving the image signal and preprocessing the image signal to obtain image data, and performing SOBEL operator calculation on a current pixel point in the image data Obtaining a SOBEL operator of the current pixel point, and correcting a SOBEL operator of the current pixel point to obtain a corrected SOBEL operator, and determining, according to the corrected SOBEL operator, whether the current pixel point is Denoising processing is required; if the current pixel point needs to perform denoising processing, calculating an average pixel value of a pixel point around the current pixel point, and using the average pixel value and the corrected SOBEL operator And performing denoising processing on the current pixel point; if the current pixel point does not need to perform denoising processing, determining whether the next pixel point in the image data needs to perform denoising processing.
  • a storage medium storing one or more modules that, when executed by an electronic device, cause the electronic device to perform the image denoising method described above.
  • the SOBEL operator can accurately distinguish the detail area and the flat area of the image, and the denoising process is not performed on the detail area, and the denoising process is performed on the flat area. Therefore, the above technical solution can preserve the image details as much as possible while improving the image signal-to-noise ratio level while the image is denoised, and the scheme is simple and easy to implement, and basically does not increase the area of the image sensor.
  • FIG. 1 is a schematic diagram of an image denoising system according to an embodiment of the present application.
  • FIG. 2 is a schematic diagram of a pixel matrix used in an image denoising system according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of a SOBEL horizontal matrix used in an image denoising system according to an embodiment of the present application
  • FIG. 4 is a schematic diagram of a SOBEL vertical matrix used in an image denoising system according to an embodiment of the present application.
  • FIG. 5 is a flowchart of an image denoising method according to an embodiment of the present application.
  • connection In the description of the present application, it should be noted that the terms “installation”, “connected”, and “connected” are to be understood broadly, and may be fixed or detachable, for example, unless otherwise specifically defined and defined. Connected, or integrally connected; may be mechanically connected, or may be electrically connected or may communicate with each other; may be directly connected or indirectly connected through an intermediate medium, may be internal communication of two elements or interaction of two elements relationship.
  • Connected, or integrally connected may be mechanically connected, or may be electrically connected or may communicate with each other; may be directly connected or indirectly connected through an intermediate medium, may be internal communication of two elements or interaction of two elements relationship.
  • the specific meanings of the above terms in the present application can be understood on a case-by-case basis.
  • an image denoising system 100 of an embodiment of the present application includes an image signal acquiring unit 102 and a processing unit 104 .
  • the image denoising system 100 of the present embodiment can be applied to CMOS (Complementary Metal Oxide Semiconductor) and CCD (Charge-coupled Device) image sensor circuits.
  • CMOS Complementary Metal Oxide Semiconductor
  • CCD Charge-coupled Device
  • the image signal acquisition unit 102 is configured to acquire and output an image signal.
  • the image signal acquisition unit 102 includes a photosensitive pixel array 106 and an image signal readout subunit 108.
  • the photosensitive pixel array 106 is for converting an optical signal into an image signal.
  • the photosensitive pixel array 106 can be a CMOS pixel array or a CCD pixel array.
  • Photosensitive pixel array 106 includes a plurality of photosensitive pixel points. The plurality of photosensitive pixel dots are arranged in two mutually perpendicular directions to form the photosensitive pixel array 106.
  • the image signal readout subunit 108 is connected to the photosensitive pixel array 106.
  • the image signal readout subunit 108 is for controlling the photosensitive pixel array 106 to convert an optical signal into an image signal and output an image signal.
  • the image signal readout subunit 108 is an integral image signal readout unit for controlling the photosensitive pixel array 106 of the current image frame to perform reset and readout line by line.
  • the processing unit 104 is configured to receive an image signal and preprocess the image signal to obtain image data, and perform SOBEL operator calculation on the current pixel point in the image data to obtain a SOBEL operator of the current pixel point, and SOBEL of the current pixel point. The operator performs correction to obtain the corrected SOBEL operator, and judges whether the current pixel point needs to be denoised according to the corrected SOBEL operator.
  • the processing unit 104 is configured to calculate an average pixel value of the pixel point around the current pixel point, and perform denoising processing on the current pixel point by using the average pixel value and the corrected SOBEL operator. .
  • the processing unit 104 It is used to determine whether the next pixel in the image data needs to be denoised.
  • the image signal output by the image signal acquiring unit 102 is generally an analog image signal
  • the processing unit 104 first digitizes the analog image signal, and then obtains the digitized image data and processes the image data. Therefore, in the present embodiment, the processing unit 104 includes an analog signal processing sub-unit 110, an analog-to-digital conversion sub-unit 112, a digital signal processing sub-unit 114, and an output sub-unit 116.
  • the analog signal processing sub-unit 110 is configured to receive an image signal output by the image signal readout subunit 108 and output the processed image signal.
  • the analog signal processing sub-unit 110 performs necessary processing on the image signal as an analog signal to acquire characteristic parameters of the image signal, which are specifically modulation, filtering, amplification, and the like.
  • the analog to digital conversion sub-unit 112 is configured to convert the image signal output by the analog signal processing sub-unit 110 into digitized image data.
  • the digitized image data is generally image data that has not been subjected to denoising processing.
  • the digital signal processing sub-unit 114 is configured to perform SOBEL operator calculation on the current pixel point in the image data to obtain the SOBEL operator of the current pixel point, and correct the SOBEL operator of the current pixel point to obtain the corrected SOBEL operator. And according to the corrected SOBEL operator, it is judged whether the current pixel needs to be denoised.
  • the digital signal processing sub-unit 114 is configured to calculate an average pixel value, and perform denoising processing on the current pixel point by using the average pixel value and the corrected SOBEL operator.
  • the digital signal processing sub-unit 114 is configured to determine whether the next pixel point in the image data needs to be denoised.
  • the digital signal processing sub-unit 114 constructs an n*n pixel matrix centering on the current pixel point, and performs a SOBEL operator calculation on the current pixel point using the n*n pixel matrix, where n is a positive odd number and n is greater than 1.
  • n 3, that is, a 3*3 pixel matrix.
  • the pixel points in the 3*3 pixel matrix are numbered as shown in FIG. 2, and the current pixel point is set to the pixel point A22 located in the middle of the pixel matrix.
  • the digital signal processing sub-unit 114 combines the SOBEL horizontal matrix and the SOBEL vertical matrix to perform SOBEL operator calculation.
  • Figure 3 shows the SOBEL horizontal matrix as an example
  • Figure 4 shows the SOBEL vertical matrix as an example.
  • the SOBEL operator of the current pixel point A22 is calculated by the following formula, and is used to measure the denoising intensity of the current pixel point:
  • C_SOBEL (A11+2*A12+A13)-(A31+2*A32+A33)+(A11+2*A21+A31)-(A13+2*A23+A33)
  • C_SOBEL is the SOBEL operator of the current pixel point A22
  • A11 represents the pixel value of the pixel point A11
  • A12 represents the pixel value of the pixel point A12
  • A13 represents the pixel value of the pixel point A13
  • A31 represents the pixel value of the pixel point A31
  • A32 The pixel value of the pixel A32 is represented
  • A33 represents the pixel value of the pixel point A33
  • A21 represents the pixel value of the pixel point A21
  • A23 represents the pixel value of the pixel point A23.
  • the digital signal processing sub-unit 114 After obtaining the SOBEL operator of the current pixel A22, the digital signal processing sub-unit 114 proceeds to the SOBEL operator. Line correction. Specifically, the digital signal processing sub-unit 114 is configured to obtain the corrected SOBEL operator according to the following formula:
  • the digital signal processing sub-unit 114 determines that the current pixel point A22 is an edge pixel point, and no denoising process is required.
  • the digital signal processing sub-unit 114 determines that the current pixel point A22 needs to be denoised.
  • a specific method of the denoising process is that the digital signal processing sub-unit 114 is configured to calculate an average pixel value of a pixel point around the current pixel point A22.
  • the digital signal processing sub-unit 114 is configured to calculate an average pixel value of a pixel point around the current pixel point A22.
  • there are eight pixels around the current pixel point A22 which are pixel points A11, A12, A13, A21, A23, A31, A32, and A33. Therefore, the average pixel value AVER is obtained according to the following formula:
  • AVER (A11+A12+A13+A21+A23+A31+A32+A33)/8.
  • the digital signal processing sub-unit 114 performs denoising processing on the current pixel point A22 according to the following formula using the average pixel value and the corrected SOBEL operator:
  • AVER_SOBEL (C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH,
  • AVER_SOBEL represents the pixel value of the current pixel point A22 after denoising
  • AVER represents the average pixel value.
  • the digital signal processing sub-unit 114 After the denoised pixel value of the current pixel point A22 is obtained, the digital signal processing sub-unit 114 overwrites the pixel value of the denoised current pixel point A22 with the current pixel point A22.
  • the digital signal processing sub-unit 114 determines whether the next pixel point in the image data needs to be denoised to complete the processing of all the pixel points of the image data. For example, the digital signal processing sub-unit 114 slides back or down in a 3*3 pixel matrix to perform a line scan of the pixel points in the image data from left to right.
  • the output subunit 116 is configured to output the image data after the denoising process.
  • the output sub-unit 116 may output the denoised image data to the storage unit 118 and/or the display unit 120 for storage and/or display purposes.
  • the pixel value of the above pixel points refers to the gray value of the pixel point in the case of brightness denoising; for example, in the case of color denoising, the R channel (red channel) of the pixel point, The average of the G channel (green channel) and B (blue channel) channels.
  • the image denoising system 100 can accurately distinguish the detail region (such as the edge) and the flat region (the pixel point of the SOBEL operator) of the image by the SOBEL operator, and does not go to the detail region. Noise processing, for flat areas, denoising processing. Therefore, the image denoising system 100 can preserve the image details as much as possible while improving the image signal-to-noise ratio level, and the image denoising system is simple and easy to implement, and basically does not increase the area of the image sensor. .
  • an embodiment of the present application provides an image denoising method.
  • the image denoising method can be implemented by the image denoising system 100 of the above embodiment.
  • the image denoising method includes:
  • S1 Acquire and output an image signal.
  • an image signal is acquired and output by an image signal acquisition unit.
  • S2 receiving the image signal and preprocessing the image signal to obtain image data, and performing SOBEL operator calculation on the current pixel point in the image data to obtain the SOBEL operator of the current pixel point, and performing the SOBEL operator of the current pixel point Correcting to obtain the corrected SOBEL operator, and determining whether the current pixel needs to be denoised according to the corrected SOBEL operator.
  • S2 can be implemented by a processing unit.
  • S4 If the current pixel point does not need to perform denoising processing, it is determined whether the next pixel point in the image data needs to be denoised.
  • S3 and S4 can be implemented by a processing unit.
  • the image signal acquisition unit includes a photosensitive pixel array and an image signal readout subunit.
  • a photosensitive pixel array is used to convert an optical signal into an image signal.
  • the photosensitive pixel array can be a CMOS pixel array or a CCD pixel array.
  • the photosensitive pixel array includes a plurality of photosensitive pixel dots. The plurality of photosensitive pixel dots are arranged in two mutually perpendicular directions to form a photosensitive pixel array.
  • the image signal readout subunit is connected to the photosensitive pixel array.
  • the image signal readout subunit is configured to control the photosensitive pixel array to convert the optical signal into an image signal and output the image signal.
  • the image signal readout subunit is an integral image signal readout unit for controlling the photosensitive pixel array of the current image frame to be reset and read out line by line.
  • the image signal output by the image signal acquisition unit is generally an analog image signal
  • the processing unit first digitizes the analog image signal, and then obtains the digitized image data and processes the image data. Therefore, in this embodiment, the processing unit includes an analog signal processing subunit, an analog to digital conversion subunit, a digital signal processing subunit, and an output subunit.
  • Performing the preprocessing includes: the analog signal processing subunit receives the image signal output by the image signal readout subunit, and outputs the processed image signal.
  • the analog to digital conversion subunit converts the image signal output by the analog signal processing subunit into digitized image data.
  • an analog signal processing subunit performs necessary for an image signal as an analog signal
  • the necessary processing is specifically modulation, filtering, and amplification.
  • the digitized image data is generally image data that has not been subjected to denoising processing.
  • the digital signal processing sub-unit performs SOBEL operator calculation on the current pixel point in the image data to obtain the SOBEL operator of the current pixel point, and corrects the SOBEL operator of the current pixel point to obtain the corrected SOBEL operator, and according to The corrected SOBEL operator determines whether the current pixel needs to be denoised.
  • the digital signal processing sub-unit constructs an n*n pixel matrix centered on the current pixel point, and performs a SOBEL operator calculation on the current pixel point using the n*n pixel matrix, where n is a positive odd number and n is greater than 1.
  • n 3, that is, a 3*3 pixel matrix.
  • the pixel points in the 3*3 pixel matrix are numbered as shown in FIG. 2, and the current pixel point is set to the pixel point A22 located in the middle of the pixel matrix.
  • the digital signal processing subunit combines the SOBEL horizontal matrix and the SOBEL vertical matrix for SOBEL operator calculation.
  • Figure 3 shows the SOBEL horizontal matrix as an example
  • Figure 4 shows the SOBEL vertical matrix as an example.
  • the SOBEL operator of the current pixel point A22 is calculated by the following formula, and is used to measure the denoising intensity of the current pixel point:
  • C_SOBEL (A11+2*A12+A13)-(A31+2*A32+A33)+(A11+2*A21+A31)-(A13+2*A23+A33)
  • C_SOBEL is the SOBEL operator of the current pixel point A22
  • A11 represents the pixel value of the pixel point A11
  • A12 represents the pixel value of the pixel point A12
  • A13 represents the pixel value of the pixel point A13
  • A31 represents the pixel value of the pixel point A31
  • A32 The pixel value of the pixel A32 is represented
  • A33 represents the pixel value of the pixel point A33
  • A21 represents the pixel value of the pixel point A21
  • A23 represents the pixel value of the pixel point A23.
  • the digital signal processing sub-unit After obtaining the SOBEL operator of the current pixel A22, the digital signal processing sub-unit corrects the SOBEL operator. Specifically, the digital signal processing subunit is configured to obtain the corrected SOBEL operator according to the following formula:
  • the digital signal processing sub-unit determines that the current pixel point A22 is an edge pixel point, and no denoising process is required, and then proceeds to S4.
  • the digital signal processing sub-unit determines that the current pixel point A22 needs to be denoised, and then proceeds to S3.
  • the digital signal processing sub-unit is used to calculate the current pixel point A22 week The average pixel value of the surrounding pixels.
  • the average pixel value AVER is obtained according to the following formula:
  • AVER (A11+A12+A13+A21+A23+A31+A32+A33)/8.
  • the digital signal processing sub-unit performs denoising processing on the current pixel point A22 according to the following formula using the average pixel value and the corrected SOBEL operator:
  • AVER_SOBEL (C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH,
  • AVER_SOBEL represents the pixel value of the current pixel point A22 after denoising
  • AVER represents the average pixel value.
  • the digital signal processing sub-unit After obtaining the pixel value of the current pixel point A22 after denoising, the digital signal processing sub-unit overwrites the current pixel point A22 with the pixel value of the denoised current pixel point A22.
  • the output subunit outputs the denoised image data.
  • the output subunit may output the denoised image data to a storage unit and/or a display unit for storage and/or display purposes.
  • the digital signal processing subunit determines whether the next pixel point in the image data needs to be denoised to complete the processing of all the pixel points of the image data. For example, the digital signal processing sub-unit slides backward or downward in a 3*3 pixel matrix to perform line scan processing of pixel points in the image data from left to right.
  • the above image denoising method can accurately distinguish the detail region (such as the edge) and the flat region (the pixel point of the SOBEL operator) of the image by the SOBEL operator, and no denoising for the detail region. Processing, for flat areas, denoising processing. Therefore, the above image denoising method can preserve the image details as much as possible while improving the image signal-to-noise ratio level, and the image denoising method is simple and easy to implement, and basically does not increase the area of the image sensor.
  • the embodiment of the present application further provides a storage medium, where the storage medium stores one or more modules, when the one or more modules are executed by one electronic device, causing the electronic device to perform any of the foregoing embodiments.
  • the image denoising method is not limited to:
  • the embodiment of the present application further provides an application program for performing the image denoising method according to any of the foregoing embodiments at runtime.
  • first and second are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated.
  • features defining “first” or “second” may include at least one of the features, either explicitly or implicitly.
  • the meaning of "a plurality” is at least two, such as two, three, etc., unless specifically defined otherwise.
  • a "computer-readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with the instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.
  • portions of the application can be implemented in hardware, software, firmware, or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals. Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.
  • each functional unit in each embodiment of the present application may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like. While the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are illustrative and are not to be construed as limiting the scope of the present application. The embodiments are subject to variations, modifications, substitutions and variations.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Facsimile Image Signal Circuits (AREA)

Abstract

公开了一种图像去噪系统和图像去噪方法和图像去噪系统，其中，图像信号获取单元获取及输出图像信号。处理单元对图像信号进行预处理得到图像数据，且对图像数据中的当前像素点进行SOBEL算子计算以得到当前像素点的SOBEL算子，并对当前像素点的SOBEL算子进行校正，并根据校正后的SOBEL算子判断当前像素点是否需进行去噪处理。如果当前像素点需进行去噪处理，则处理单元计算当前像素点周围的像素点的平均像素值，并利用平均像素值及校正后的SOBEL算子对当前像素点进行去噪处理。如果当前像素点无需进行去噪处理，则处理单元判断图像数据中的下一个像素点是否需进行去噪处理。

Description

图像去噪系统及图像去噪方法
相关申请的交叉引用
本申请要求中国专利申请号201510505819.8、申请日为2015年8月17日的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及图像处理领域，更具体而言，涉及一种图像去噪系统及一种图像去噪方法。
背景技术
近年来图像传感器发展迅速，在销量不断攀升的同时对于图像的质量要求也越来越高，其中最重要的指标之一就是信噪比。受硬件、环境和人为的影响，图像中必然会存有噪声及各种杂讯，从而很大程度上影响了图像的细节，最终影响图像质量。相应的，各种去噪方法陆续涌现出来，如局部自适应去噪、基于方差去噪、小波变换去噪、中值滤波去噪等。
如前所述，现有的图像去噪方法包括局部自适应去噪、基于方差去噪、小波变换去噪、中值滤波去噪等。去噪时要注意，一方面由于噪声处在高频区域，容易和图像细节混淆，在去噪的同时带来细节的损失，使得图像变得模糊；另一方面，某些去噪算法(如小波边缘去噪)运算量太大，实现较复杂，不适合用于近年来低成本设计的CMOS和CCD等图像传感器电路中。
发明内容
本申请旨在至少解决现有技术中存在的技术问题之一。
为此，本申请提供了一种图像去噪系统及一种图像去噪方法。
一种图像去噪系统，包括：图像信号获取单元，所述图像信号获取单元用于获取及输出图像信号；及处理单元，所述处理单元用于接收所述图像信号并对所述图像信号进行预处理得到图像数据，且对所述图像数据中的当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，并对所述当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理；所述处理单元还用于在判断所述当前像素点需进行去噪处理时，计算所述当前像素点周围的像素点的平均像素值，并利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理；所述处理单元还用于在判断所述当前像素点无需进行去噪处理时，判 断所述图像数据中的下一个像素点是否需进行去噪处理。
一种图像去噪方法，包括：获取及输出图像信号；接收所述图像信号并对所述图像信号进行预处理得到图像数据，且对所述图像数据中的当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，并对所述当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理；如果所述当前像素点需进行去噪处理，则计算所述当前像素点周围的像素点的平均像素值，并利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理；如果所述当前像素点无需进行去噪处理，则判断所述图像数据中的下一个像素点是否需进行去噪处理。
一种存储介质，存储有一个或者多个模块，当所述一个或者多个模块被一个电子设备执行时，使得所述电子设备执行上述的图像去噪方法。
一种应用程序，用于在运行时，执行上述的图像去噪方法。
上述技术方案，通过SOBEL算子能比较准确地区分图像的细节区域和平坦区域，对于细节区域，不进行去噪处理，对于平坦区域，进行去噪处理。因此上述技术方案在对图像进行去噪处理的同时能够尽可能地保留图像细节，提高图像信噪比水平，而且该方案简单易于实现，基本上不会增加图像传感器的面积。
本申请的附加方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实践了解到。
附图说明
本申请的上述和/或附加的方面和优点从结合下面附图对实施方式的描述中将变得明显和容易理解，其中：
图1是本申请实施方式的图像去噪系统的示意图；
图2是本申请实施方式的图像去噪系统所用的像素矩阵的示意图；
图3是本申请实施方式的图像去噪系统所用的SOBEL水平矩阵的示意图；
图4是本申请实施方式的图像去噪系统所用的SOBEL垂直矩阵的示意图；及
图5是本申请实施方式的图像去噪方法的流程图。
具体实施方式
下面详细描述本申请的实施方式，所述实施方式的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附 图描述的实施方式是示例性的，仅用于解释本申请，而不能理解为对本申请的限制。
在本申请的描述中，需要说明的是，除非另有明确的规定和限定，术语“安装”、“相连”、“连接”应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或一体地连接；可以是机械连接，也可以是电连接或可以相互通信；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通或两个元件的相互作用关系。对于本领域的普通技术人员而言，可以根据具体情况理解上述术语在本申请中的具体含义。
下文的公开提供了许多不同的实施方式或例子用来实现本申请的不同结构。为了简化本申请的公开，下文中对特定例子的部件和设定进行描述。当然，它们仅仅为示例，并且目的不在于限制本申请。此外，本申请可以在不同例子中重复参考数字和/或参考字母，这种重复是为了简化和清楚的目的，其本身不指示所讨论各种实施方式和/或设定之间的关系。此外，本申请提供了的各种特定的工艺和材料的例子，但是本领域普通技术人员可以意识到其他工艺的应用和/或其他材料的使用。
请参阅图1，本申请实施方式的图像去噪系统100包括图像信号获取单元102及处理单元104。本实施方式的图像去噪系统100可适用于CMOS(Complementary Metal Oxide Semiconductor，互补金属氧化物半导体)和CCD(Charge-coupled Device，电荷耦合元件)图像传感器电路中。
图像信号获取单元102用于获取及输出图像信号。具体地，本实施方式中，图像信号获取单元102包括感光像素阵列106及图像信号读出子单元108。
感光像素阵列106用于将光信号转换为图像信号。例如，感光像素阵列106可为CMOS像素阵列或CCD像素阵列。感光像素阵列106包括多个感光像素点。多个感光像素点沿相互垂直的两个方向排列以形成感光像素阵列106。
图像信号读出子单元108连接感光像素阵列106。图像信号读出子单元108用于控制感光像素阵列106将光信号转换为图像信号，并输出图像信号。例如，图像信号读出子单元108为积分图像信号读出单元，用于控制当前图像帧的感光像素阵列106逐行进行复位和读出。
处理单元104用于接收图像信号并对图像信号进行预处理得到图像数据，且对图像数据中的当前像素点进行SOBEL算子计算以得到当前像素点的SOBEL算子，并对当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据校正后的SOBEL算子判断当前像素点是否需进行去噪处理。在判断当前像素点需进行去噪处理时，处理单元104用于计算当前像素点周围的像素点的平均像素值，并利用平均像素值及校正后的SOBEL算子对当前像素点进行去噪处理。在判断当前像素点无需进行去噪处理时，处理单元104 用于判断图像数据中的下一个像素点是否需进行去噪处理。
具体地，图像信号获取单元102输出的图像信号一般为模拟图像信号，处理单元104先要对模拟图像信号数字化，然后得到数字化后的图像数据并对图像数据进行处理。因此，本实施方式中，处理单元104包括模拟信号处理子单元110、模数转换子单元112、数字信号处理子单元114及输出子单元116。
模拟信号处理子单元110用于接收图像信号读出子单元108输出的图像信号，并输出经处理后的图像信号。例如，模拟信号处理子单元110对作为模拟信号的图像信号进行必要的处理以获取图像信号的特征参数，所述必要的处理具体为调制、滤波及放大等。
模数转换子单元112用于将模拟信号处理子单元110输出的图像信号转换为数字化的图像数据。数字化后的图像数据一般是未进行去噪处理的图像数据。
数字信号处理子单元114用于对图像数据中的当前像素点进行SOBEL算子计算以得到当前像素点的SOBEL算子，并对当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据校正后的SOBEL算子判断当前像素点是否需进行去噪处理。在当前像素点需进行去噪处理时，数字信号处理子单元114用于计算平均像素值，并利用平均像素值及校正后的SOBEL算子对当前像素点进行去噪处理。在当前像素点无需进行去噪处理时，数字信号处理子单元114用于判断图像数据中的下一个像素点是否需进行去噪处理。
具体地，数字信号处理子单元114以当前像素点为中心，构造n*n像素矩阵，并利用n*n像素矩阵对当前像素点进行SOBEL算子计算，其中n为正奇数且n大于1。本实施方式中，n＝3，即3*3像素矩阵。为方便说明，对3*3像素矩阵中的像素点进行编号，如图2所示，并将当前像素点设为位于像素矩阵中间的像素点A22。
同时，数字信号处理子单元114结合SOBEL水平矩阵及SOBEL垂直矩阵进行SOBEL算子计算。图3显示作为一个例子的SOBEL水平矩阵，图4显示作为一个例子的SOBEL垂直矩阵。
因此，当前像素点A22的SOBEL算子由以下公式计算，并以此来衡量当前像素点的去噪强度大小：
C_SOBEL＝(A11+2*A12+A13)-(A31+2*A32+A33)+(A11+2*A21+A31)-(A13+2*A23+A33)
其中，C_SOBEL为当前像素点A22的SOBEL算子，A11表示像素点A11的像素值，A12表示像素点A12的像素值，A13表示像素点A13的像素值，A31表示像素点A31的像素值，A32表示像素点A32的像素值，A33表示像素点A33的像素值，A21表示像素点A21的像素值，A23表示像素点A23的像素值。
得到当前像素点A22的SOBEL算子之后，数字信号处理子单元114对SOBEL算子进 行校正。具体地，数字信号处理子单元114用于根据以下公式获取校正后的SOBEL算子：
Figure PCTCN2016094817-appb-000001
其中，TH为设定值，可根据经验值确定，例如，TH＝96，C_SOBEL_T为校正后的SOBEL算子，C_SOBEL为校正前的SOBEL算子。
对于C_SOBEL_T＝TH的当前像素点A22，数字信号处理子单元114判断当前像素点A22为边缘像素点，可无需进行去噪处理。
对于C_SOBEL_T＝C_SOBEL的当前像素点A22，数字信号处理子单元114判断当前像素点A22需进行去噪处理。
去噪处理的具体方法是：数字信号处理子单元114用于计算当前像素点A22周围的像素点的平均像素值。本实施方式中，当前像素点A22周围的像素点有8个，分别为像素点A11、A12、A13、A21、A23、A31、A32及A33，因此，平均像素值AVER根据以下公式获得：
AVER＝(A11+A12+A13+A21+A23+A31+A32+A33)/8。
数字信号处理子单元114利用平均像素值及校正后的SOBEL算子根据以下公式对当前像素点A22进行去噪处理：
AVER_SOBEL＝(C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH，
其中，AVER_SOBEL表示去噪后的当前像素点A22的像素值，A表示去噪前的当前像素点A22的像素值，例如，A＝A22，AVER表示平均像素值。
在得到去噪后的当前像素点A22的像素值后，数字信号处理子单元114将去噪后的当前像素点A22的像素值覆盖当前像素点A22。
若当前像素点A22无需进行去噪处理，数字信号处理子单元114对图像数据中的下一像素点判断是否需进行去噪处理以完成图像数据的所有像素点的处理。例如，数字信号处理子单元114以3*3像素矩阵向后或向下滑动以对图像数据中的像素点从左至右进行行扫描一一处理。
输出子单元116用于输出经去噪处理后的图像数据。例如，输出子单元116可将经去噪处理后的图像数据输出至存储单元118及/或显示单元120以作存储及/或显示的用途。
需要指出的是，上述像素点的像素值如在亮度去噪的情况下，是指像素点的灰度值；如在色彩去噪的情况下，是指像素点的R通道(红色通道)、G通道(绿色通道)和B(蓝色通道)通道的平均值。
综上所述，上述图像去噪系统100通过SOBEL算子能比较准确地区分图像的细节区域(如边缘)和平坦区域(SOBEL算子较小的像素点)，对于细节区域处，不进行去噪处理，对于平坦区域，进行去噪处理。因此上述图像去噪系统100在对图像进行去噪处理的同时能够尽可能地保留图像细节，提高图像信噪比水平，而且该图像去噪系统简单易于实现，基本上不会增加图像传感器的面积。
请参图5，本申请实施方式提供一种图像去噪方法。该图像去噪方法可由以上实施方式的图像去噪系统100实现。该图像去噪方法包括：
S1：获取及输出图像信号。例如，通过图像信号获取单元获取及输出图像信号。
S2：接收图像信号并对图像信号进行预处理得到图像数据，且对图像数据中的当前像素点进行SOBEL算子计算以得到当前像素点的SOBEL算子，并对当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据校正后的SOBEL算子判断当前像素点是否需进行去噪处理。例如，S2可由处理单元实现。
S3：如果当前像素点需进行去噪处理，则计算当前像素点周围的像素点的平均像素值，并利用平均像素值及校正后的SOBEL算子对当前像素点进行去噪处理。
S4：如果当前像素点无需进行去噪处理，则判断图像数据中的下一个像素点是否需进行去噪处理。例如，S3和S4可由处理单元实现。
在S1中，具体地，本实施方式中，图像信号获取单元包括感光像素阵列及图像信号读出子单元。
感光像素阵列用于将光信号转换为图像信号。例如，感光像素阵列可为CMOS像素阵列或CCD像素阵列。感光像素阵列包括多个感光像素点。多个感光像素点沿相互垂直的两个方向排列以形成感光像素阵列。
图像信号读出子单元连接感光像素阵列。图像信号读出子单元用于控制感光像素阵列将光信号转换为图像信号，并输出图像信号。例如，图像信号读出子单元为积分图像信号读出单元，用于控制当前图像帧的感光像素阵列逐行进行复位和读出。
在S2中，图像信号获取单元输出的图像信号一般为模拟图像信号，处理单元先要对模拟图像信号数字化，然后得到数字化后的图像数据并对图像数据进行处理。因此，本实施方式中，处理单元包括模拟信号处理子单元、模数转换子单元、数字信号处理子单元及输出子单元。
进行预处理包括：模拟信号处理子单元接收图像信号读出子单元输出的图像信号，并输出经处理后的图像信号。模数转换子单元将模拟信号处理子单元输出的图像信号转换为数字化的图像数据。例如，模拟信号处理子单元对作为模拟信号的图像信号进行必要的处 理以获取图像信号的特征参数，所述必要的处理具体为调制、滤波及放大等。数字化后的图像数据一般是未进行去噪处理的图像数据。
数字信号处理子单元对图像数据中的当前像素点进行SOBEL算子计算以得到当前像素点的SOBEL算子，并对当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据校正后的SOBEL算子判断当前像素点是否需进行去噪处理。
具体地，数字信号处理子单元以当前像素点为中心，构造n*n像素矩阵，并利用n*n像素矩阵对当前像素点进行SOBEL算子计算，其中n为正奇数且n大于1。本实施方式中，n＝3，即3*3像素矩阵。为方便说明，对3*3像素矩阵中的像素点进行编号，如图2所示，并将当前像素点设为位于像素矩阵中间的像素点A22。
同时，数字信号处理子单元结合SOBEL水平矩阵及SOBEL垂直矩阵进行SOBEL算子计算。图3显示作为一个例子的SOBEL水平矩阵，图4显示作为一个例子的SOBEL垂直矩阵。
因此，当前像素点A22的SOBEL算子由以下公式计算，并以此来衡量当前像素点的去噪强度大小：
C_SOBEL＝(A11+2*A12+A13)-(A31+2*A32+A33)+(A11+2*A21+A31)-(A13+2*A23+A33)
其中，C_SOBEL为当前像素点A22的SOBEL算子，A11表示像素点A11的像素值，A12表示像素点A12的像素值，A13表示像素点A13的像素值，A31表示像素点A31的像素值，A32表示像素点A32的像素值，A33表示像素点A33的像素值，A21表示像素点A21的像素值，A23表示像素点A23的像素值。
得到当前像素点A22的SOBEL算子之后，数字信号处理子单元对SOBEL算子进行校正。具体地，数字信号处理子单元用于根据以下公式获取校正后的SOBEL算子：
Figure PCTCN2016094817-appb-000002
其中，TH为设定值，可根据经验值确定，例如，TH＝96，C_SOBEL_T为校正后的SOBEL算子，C_SOBEL为校正前的SOBEL算子。
对于C_SOBEL_T＝TH的当前像素点A22，数字信号处理子单元判断当前像素点A22为边缘像素点，可无需进行去噪处理，之后进入S4。
对于C_SOBEL_T＝C_SOBEL的当前像素点A22，数字信号处理子单元判断当前像素点A22需进行去噪处理，之后进入S3。
在S3中，去噪处理的具体方法是：数字信号处理子单元用于计算当前像素点A22周 围的像素点的平均像素值。本实施方式中，当前像素点A22周围的像素点有8个，分别为像素点A11、A12、A13、A21、A23、A31、A32及A33，因此，平均像素值AVER根据以下公式获得：
AVER＝(A11+A12+A13+A21+A23+A31+A32+A33)/8。
数字信号处理子单元利用平均像素值及校正后的SOBEL算子根据以下公式对当前像素点A22进行去噪处理：
AVER_SOBEL＝(C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH，
其中，AVER_SOBEL表示去噪后的当前像素点A22的像素值，A表示去噪前的当前像素点A22的像素值，例如，A＝A22，AVER表示平均像素值。
在得到去噪后的当前像素点A22的像素值后，数字信号处理子单元将去噪后的当前像素点A22的像素值覆盖当前像素点A22。
输出子单元输出经去噪处理后的图像数据。例如，输出子单元可将经去噪处理后的图像数据输出至存储单元及/或显示单元以作存储及/或显示的用途。
在S4中，即当前像素点A22无需进行去噪处理时，数字信号处理子单元对图像数据中的下一像素点判断是否需进行去噪处理以完成图像数据的所有像素点的处理。例如，数字信号处理子单元以3*3像素矩阵向后或向下滑动以对图像数据中的像素点从左至右进行行扫描一一处理。
综上所述，上述图像去噪方法通过SOBEL算子能比较准确地区分图像的细节区域(如边缘)和平坦区域(SOBEL算子较小的像素点)，对于细节区域处，不进行去噪处理，对于平坦区域，进行去噪处理。因此上述图像去噪方法在对图像进行去噪处理的同时能够尽可能地保留图像细节，提高图像信噪比水平，而且该图像去噪方法简单易于实现，基本上不会增加图像传感器的面积。
本申请实施例还提供了一种存储介质，所述存储介质存储有一个或者多个模块，当所述一个或者多个模块被一个电子设备执行时，使得所述电子设备执行前述任一实施例所述的图像去噪方法。
本申请实施例还提供了一种应用程序，用于在运行时，执行前述任一实施例所述的图像去噪方法。
在本说明书的描述中，参考术语“一个实施方式”、“一些实施方式”、“示意性实施方式”、“示例”、“具体示例”、或“一些示例”等的描述意指结合所述实施方式或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施方式或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施方式或示例。而且，描述的具体特征、结构、 材料或者特点可以在任何的一个或多个实施方式或示例中以合适的方式结合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本申请的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本申请的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中， 该程序在执行时，包括方法实施例的步骤之一或其组合。
此外，在本申请各个实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本申请的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (14)

  1. 一种图像去噪系统，其特征在于，包括：
    图像信号获取单元，所述图像信号获取单元用于获取及输出图像信号；及
    处理单元，所述处理单元用于接收所述图像信号并对所述图像信号进行预处理得到图像数据，且对所述图像数据中的当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，并对所述当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理；
    所述处理单元还用于在判断所述当前像素点需进行去噪处理时，计算所述当前像素点周围的像素点的平均像素值，并利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理；
    所述处理单元还用于在判断所述当前像素点无需进行去噪处理时，判断所述图像数据中的下一个像素点是否需进行去噪处理。
  2. 如权利要求1所述的图像去噪系统，其特征在于，所述处理单元还用于以所述当前像素点为中心，构造n*n像素矩阵，并利用所述n*n像素矩阵对所述当前像素点进行SOBEL算子计算，其中n为正奇数且n大于1。
  3. 如权利要求1或2所述的图像去噪系统，其特征在于，所述处理单元还用于根据以下公式获取所述校正后的SOBEL算子：
    Figure PCTCN2016094817-appb-100001
    其中，TH为设定值，C_SOBEL_T为所述校正后的SOBEL算子，C_SOBEL为校正前的所述SOBEL算子。
  4. 如权利要求3所述的图像去噪系统，其特征在于，所述处理单元还用于：
    在所述校正后的SOBEL算子等于所述设定值时，确定所述当前像素点无需进行去噪处理；
    在所述校正后的SOBEL算子等于校正前的所述SOBEL算子时，确定所述当前像素点需进行去噪处理。
  5. 如权利要求1至4中任一项所述的图像去噪系统，其特征在于，所述处理单元用于根据以下公式对所述当前像素点进行去噪处理：
    AVER_SOBEL＝(C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH，
    其中，AVER_SOBEL表示去噪后的所述当前像素点的像素值，A表示去噪前的所述当前像素点的像素值，C_SOBEL_T为所述校正后的SOBEL算子，TH为设定值，AVER表示所述平均像素值。
  6. 如权利要求1至5中任一项所述的图像去噪系统，其特征在于，所述图像信号获取单元包括：
    感光像素阵列，所述感光像素阵列用于将光信号转换为所述图像信号；及
    图像信号读出子单元，所述图像信号读出单元用于控制所述感光像素阵列将所述光信号转换为所述图像信号，并输出所述图像信号。
  7. 如权利要求6所述的图像去噪系统，其特征在于，所述处理单元包括：
    模拟信号处理子单元，所述模拟信号处理子单元用于接收所述图像信号读出单元输出的所述图像信号，并输出经处理后的所述图像信号；
    模数转换子单元，所述模数转换子单元用于将所述模拟信号处理子单元输出的所述图像信号转换为数字化的所述图像数据；
    数字信号处理子单元，所述数字信号处理子单元用于对所述图像数据中的所述当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，并对所述当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理；
    所述数字信号处理子单元还用于在所述当前像素点需进行去噪处理时，计算所述平均像素值，并利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理；
    所述数字信号处理子单元还用于在所述当前像素点无需进行去噪处理时，判断所述图像数据中的下一个像素点是否需进行去噪处理；及
    输出子单元，所述输出子单元用于输出经去噪处理后的所述图像数据。
  8. 一种图像去噪方法，其特征在于，包括：
    获取及输出图像信号；
    接收所述图像信号并对所述图像信号进行预处理得到图像数据，且对所述图像数据中 的当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，并对所述当前像素点的SOBEL算子进行校正以获得校正后的SOBEL算子，并根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理；
    如果所述当前像素点需进行去噪处理，则计算所述当前像素点周围的像素点的平均像素值，并利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理；
    如果所述当前像素点无需进行去噪处理，则判断所述图像数据中的下一个像素点是否需进行去噪处理。
  9. 如权利要求8所述的图像去噪方法，其特征在于，所述对所述图像数据中的当前像素点进行SOBEL算子计算以得到所述当前像素点的SOBEL算子，包括：
    以所述当前像素点为中心，构造n*n像素矩阵，并利用所述n*n像素矩阵对所述当前像素点进行SOBEL算子计算，其中n为正奇数且n大于1。
  10. 如权利要求8或9所述的图像去噪方法，其特征在于，所述对所述当前像素点的SOBEL算子进行校正以获得所述校正后的SOBEL算子，包括：
    根据以下公式获取所述校正后的SOBEL算子：
    Figure PCTCN2016094817-appb-100002
    其中，TH为设定值，C_SOBEL_T为所述校正后的SOBEL算子，C_SOBEL为校正前的所述SOBEL算子。
  11. 如权利要求10所述的图像去噪方法，其特征在于，所述根据所述校正后的SOBEL算子判断所述当前像素点是否需进行去噪处理，包括：
    如果所述校正后的SOBEL算子等于所述设定值，则所述处理单元判断所述当前像素点无需进行去噪处理；
    如果所述校正后的SOBEL算子等于校正前的所述SOBEL算子，则所述处理单元判断所述当前像素点需进行去噪处理。
  12. 如权利要求8至11中任一项所述的图像去噪方法，其特征在于，所述利用所述平均像素值及所述校正后的SOBEL算子对所述当前像素点进行去噪处理，包括：
    根据以下公式对所述当前像素点进行去噪处理：
    AVER_SOBEL＝(C_SOBEL_T*A)/TH+(TH-C_SOBEL_T)*AVER/TH，
    其中，AVER_SOBEL表示去噪后的所述当前像素点的像素值，A表示去噪前的所述当前像素点的像素值，C_SOBEL_T为所述校正后的SOBEL算子，TH为设定值，AVER表示所述平均像素值。
  13. 一种存储介质，其特征在于，所述存储介质存储有一个或者多个模块，当所述一个或者多个模块被一个电子设备执行时，使得所述电子设备执行：
    权利要求8至12中任一项所述的图像去噪方法。
  14. 一种应用程序，其特征在于，用于在运行时，执行权利要求8至12中任一项所述的图像去噪方法。
PCT/CN2016/094817 2015-08-17 2016-08-12 图像去噪系统及图像去噪方法 Ceased WO2017028742A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
KR1020187003980A KR20180033211A (ko) 2015-08-17 2016-08-12 이미지 노이즈 제거 시스템 및 이미지 노이즈 제거 방법
US15/748,699 US20190005624A1 (en) 2015-08-17 2016-08-12 Image denoising system and image denoising method
EP16836606.0A EP3340168A4 (en) 2015-08-17 2016-08-12 PICTURE COUNTERYSTEM AND PICTURE EXAMINATION METHOD

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510505819.8A CN106469436B (zh) 2015-08-17 2015-08-17 图像去噪系统及图像去噪方法
CN201510505819.8 2015-08-17

Publications (1)

Publication Number Publication Date
WO2017028742A1 true WO2017028742A1 (zh) 2017-02-23

Family

ID=58050784

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/094817 Ceased WO2017028742A1 (zh) 2015-08-17 2016-08-12 图像去噪系统及图像去噪方法

Country Status (5)

Country Link
US (1) US20190005624A1 (zh)
EP (1) EP3340168A4 (zh)
KR (1) KR20180033211A (zh)
CN (1) CN106469436B (zh)
WO (1) WO2017028742A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109242803A (zh) * 2018-10-10 2019-01-18 北京航天易联科技发展有限公司 一种应用于微波图像的复原方法及装置
CN114519675A (zh) * 2022-01-21 2022-05-20 百果园技术(新加坡)有限公司 图像处理方法、装置、电子设备及可读存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108270942B (zh) * 2018-01-31 2020-09-25 威海华菱光电股份有限公司 图像扫描装置、控制图像扫描光信号的接收方法及装置
CN108460745A (zh) * 2018-03-29 2018-08-28 哈尔滨理工大学 一种基于非局部均值滤波的图像去噪方法
KR102614088B1 (ko) * 2018-08-06 2023-12-18 삼성전자주식회사 이미지 신호 프로세서 및 이미지 신호 프로세서를 포함하는 전자 장치
CN111289848B (zh) * 2020-01-13 2023-04-07 甘肃省安全生产科学研究院有限公司 一种应用在基于安全生产的智能型热局放仪的复合数据滤波方法
CN111583145B (zh) * 2020-04-30 2023-11-28 深圳市商汤智能传感科技有限公司 图像降噪方法及装置、电子设备和存储介质
CN113379640B (zh) * 2021-06-25 2023-06-27 哈尔滨工业大学 一种融合边缘信息的多级滤波图像去噪方法
CN116309130B (zh) * 2023-02-16 2026-03-10 维沃移动通信有限公司 图像处理方法及其装置
CN116188312B (zh) * 2023-03-03 2025-11-25 杭州微引科技有限公司 一种超声图像去噪方法、电子设备及存储介质
CN116452433B (zh) * 2023-03-10 2025-10-03 浙江大学 用于光存储点阵读取的基于Sobel滤波的检测方法
CN118365555B (zh) * 2024-06-20 2024-08-13 厦门海荭兴仪器股份有限公司 基于深度学习的光谱背景噪声和干扰消除系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1633159A (zh) * 2005-01-26 2005-06-29 北京中星微电子有限公司 一种去除图像噪声的方法
US20060018564A1 (en) * 2004-07-21 2006-01-26 Kabushiki Kaisha Toshiba Interpolation pixel generation circuit
JP2008210387A (ja) * 2007-02-26 2008-09-11 Fujitsu Ltd 文書画像の２値化性能を改善するノイズ除去装置及びノイズ除去プログラム
CN103020911A (zh) * 2012-12-12 2013-04-03 中国航空工业集团公司洛阳电光设备研究所 一种用于去除混合噪声的图像处理方法
CN104657735A (zh) * 2013-11-21 2015-05-27 比亚迪股份有限公司 车道线检测方法、系统、车道偏离预警方法及系统
CN104778669A (zh) * 2015-04-16 2015-07-15 北京邮电大学 快速图像去噪方法及装置
CN104794685A (zh) * 2014-01-17 2015-07-22 腾讯科技（深圳）有限公司 一种实现图像去噪的方法及装置

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6229578B1 (en) * 1997-12-08 2001-05-08 Intel Corporation Edge-detection based noise removal algorithm
US20090251570A1 (en) * 2008-04-02 2009-10-08 Miaohong Shi Apparatus and method for noise reduction
US8204328B2 (en) * 2008-04-08 2012-06-19 The United States Of America, As Represented By The Secretary Of The Navy Automated underwater image restoration via denoised deconvolution
US8731337B2 (en) * 2011-08-05 2014-05-20 Adobe Systems Incorporated Denoising and artifact removal in image upscaling
CN102999888B (zh) * 2012-11-27 2015-02-25 西安交通大学 一种基于彩色图像分割的深度图去噪方法
CN103606132B (zh) * 2013-10-31 2016-04-13 西安电子科技大学 基于空域和时域联合滤波的多帧数字图像去噪方法
CN103632352B (zh) * 2013-11-01 2017-04-26 华为技术有限公司 一种噪声图像的时域降噪方法和相关装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060018564A1 (en) * 2004-07-21 2006-01-26 Kabushiki Kaisha Toshiba Interpolation pixel generation circuit
CN1633159A (zh) * 2005-01-26 2005-06-29 北京中星微电子有限公司 一种去除图像噪声的方法
JP2008210387A (ja) * 2007-02-26 2008-09-11 Fujitsu Ltd 文書画像の２値化性能を改善するノイズ除去装置及びノイズ除去プログラム
CN103020911A (zh) * 2012-12-12 2013-04-03 中国航空工业集团公司洛阳电光设备研究所 一种用于去除混合噪声的图像处理方法
CN104657735A (zh) * 2013-11-21 2015-05-27 比亚迪股份有限公司 车道线检测方法、系统、车道偏离预警方法及系统
CN104794685A (zh) * 2014-01-17 2015-07-22 腾讯科技（深圳）有限公司 一种实现图像去噪的方法及装置
CN104778669A (zh) * 2015-04-16 2015-07-15 北京邮电大学 快速图像去噪方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3340168A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109242803A (zh) * 2018-10-10 2019-01-18 北京航天易联科技发展有限公司 一种应用于微波图像的复原方法及装置
CN114519675A (zh) * 2022-01-21 2022-05-20 百果园技术(新加坡)有限公司 图像处理方法、装置、电子设备及可读存储介质

Also Published As

Publication number Publication date
US20190005624A1 (en) 2019-01-03
EP3340168A4 (en) 2019-02-27
CN106469436B (zh) 2019-11-08
EP3340168A1 (en) 2018-06-27
KR20180033211A (ko) 2018-04-02
CN106469436A (zh) 2017-03-01

Similar Documents

Publication Publication Date Title
CN106469436B (zh) 图像去噪系统及图像去噪方法
EP2278788B1 (en) Method and apparatus for correcting lens shading
JP5904213B2 (ja) 画像処理装置、および画像処理方法、並びにプログラム
JP4885179B2 (ja) 映像歪曲補正方法及びその補正方法を採用した映像処理装置
CN114359055B (zh) 一种多相机拍摄屏体的图像拼接方法及相关装置
CN107071234B (zh) 一种镜头阴影校正方法及装置
US10058237B2 (en) Image processing device, image processing method, and program
JP2021122106A (ja) 撮像装置、学習装置、撮像装置の制御方法、学習方法、学習済みモデルおよびプログラム
EP1976308B1 (en) Device and method for measuring noise characteristics of image sensor
EP2677494B1 (en) Image processing apparatus and image processing method
KR101578029B1 (ko) 영상 왜곡 보정 장치 및 방법
CN113409366B (zh) 一种遥感卫星影像的辐射校正方法及装置
US11228724B2 (en) Image processing apparatus, image processing method, and computer-readable recording medium for interpolating pixel value of determined defective pixel to be corrected
WO2025011130A1 (en) Image acquisition systems and methods
JP6478774B2 (ja) 撮像装置、撮像方法及びプログラム
JP2017017637A (ja) 画像処理装置及び画像処理方法
JP5167614B2 (ja) 距離画像生成装置、距離画像生成方法及びプログラム
JP2011228855A (ja) 画像処理装置、画像ノイズ判定方法、およびプログラム
JP2017050700A (ja) 画像処理装置、撮像装置および画像処理プログラム
CN106920217B (zh) 图像矫正的方法及装置
JP6251272B2 (ja) 固定パターンノイズ低減
CN115665343A (zh) 图像处理方法和可读存储介质
JP2016195367A (ja) 画像処理装置及び方法、撮像装置、並びにプログラム
CN104937916A (zh) 图像校正系统、图像校正方法和程序记录介质
JP2007228259A (ja) 信号処理装置及び撮像装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16836606

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20187003980

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2016836606

Country of ref document: EP