WO2023022630A1 - Способ измерения линейных размеров объекта по изображению - Google Patents
Способ измерения линейных размеров объекта по изображению Download PDFInfo
- Publication number
- WO2023022630A1 WO2023022630A1 PCT/RU2022/050252 RU2022050252W WO2023022630A1 WO 2023022630 A1 WO2023022630 A1 WO 2023022630A1 RU 2022050252 W RU2022050252 W RU 2022050252W WO 2023022630 A1 WO2023022630 A1 WO 2023022630A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- dimensions
- measured
- contour
- image
- determining
- 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
Links
Classifications
-
- A—HUMAN NECESSITIES
- A41—WEARING APPAREL
- A41H—APPLIANCES OR METHODS FOR MAKING CLOTHES, e.g. FOR DRESS-MAKING OR FOR TAILORING, NOT OTHERWISE PROVIDED FOR
- A41H1/00—Measuring aids or methods
- A41H1/02—Devices for taking measurements on the human body
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/88—Image or video recognition using optical means, e.g. reference filters, holographic masks, frequency domain filters or spatial domain filters
- G06V10/92—Image or video recognition using optical means, e.g. reference filters, holographic masks, frequency domain filters or spatial domain filters using spatial domain filters, e.g. joint transform correlators
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/12—Acquisition of 3D measurements of objects
Definitions
- the invention relates to methods for non-contact determination of the linear dimensions of an object and can be used to determine the anthropometric dimensions of human body parts in the virtual selection and ordering of clothing in online services and its production.
- JP2017101356, IPC: A41H1/02, published 08/06/2017 taken as a prototype, in which a set of user photographs is produced, including at least one image of the user from the front or back, as well as a side view .
- the actual dimensions are determined, which serve as a guide for the dimensions in the above images.
- every part of the user's body can be measured from a simple shooting image, which greatly reduces the user's time in making clothes.
- the measurement does not require special knowledge or skills, nor special or expensive tools.
- the disadvantage of this known method is that this method involves finding measurement points on the image with the participation of a person, the measurement points can be located in different planes relative to each other, which can reduce the accuracy of measurements due to errors introduced during manual processing of photographs.
- the participation of a second person or special equipment for surveying is required to carry out the measurements .
- the technical result achieved by the claimed method consists in increasing the accuracy of non-contact determination of linear dimensions, increasing the speed of obtaining the measurement result through the use of computer vision algorithms and neural network technologies that analyze images without participation a person in the process of determining the size of the measured object, which eliminates the occurrence of errors that can be caused by human inattention.
- the technical result is achieved due to the fact that in order to obtain an image in the frame, an exemplary object is placed along with the measured object, photographs are taken from different angles using a mobile device, for example, from above and from the side to determine the width and length of the object, respectively, and they are transmitted electronically to the server into a computational algorithm, then sequentially on each image using computer vision algorithms and an artificial neural network, the contours of the measured and reference objects are determined, at which the coordinates of the extreme points of the contour of the exemplary object are found and a projective transformation is performed to find the actual dimensions of the measured object, taking into account perspective distortions of the frame.
- a projective transformation is a formula that allows you to convert the apparent dimensions of an object in a frame to its true dimensions, taking into account the fact that objects further away have smaller apparent dimensions.
- the projective transformation eliminates the distortion of the visualization of the size of the object, which occurs when the camera is located at an arbitrary angle to the object when shooting.
- the claimed method is implemented sequentially. Using a mobile phone with software installed on it, an image is captured so that the measured and exemplary objects fall into the frame, photographs are taken from different angles, for example, from above and from the side to obtain the width and length of the object, respectively. The resulting images are transmitted to the server, where they are sequentially processed.
- Image processing is carried out by computer vision methods using an artificial neural network, while finding the pixel mask and contour of the exemplary object, approximating the contour with a polygon, obtaining corner points, then finding the pixel mask and contour of the measured object, determining the extreme points of the contour of the exemplary object of known size and shape, after which, using a projective transformation, the physical dimensions of the measured object are found, taking into account perspective distortions caused by an arbitrary position of the camera when shooting.
- the result of the implementation of the method is one size of the measured object (length or width) for one source image.
- the measurement result is displayed on the mobile phone of the user who took the photos.
- measuring the linear dimensions of an object involves creating a photograph or a set of photographs of the object being measured in different projections, while simultaneously placing in the frame an exemplary object ( landmark object) with pre-known dimensions, placed so that the dimensions can be count from photographs, image processing by a computational algorithm, calculation of the actual size of the measured object.
- the claimed method includes the detection of contours and characteristic points (for example, corners) of a model object. Further, taking into account the previously known shape of the exemplary object (rectangle, circle, etc.), information about which is stored in the database of exemplary objects on the server, a projective transformation is built, which allows compensating for distortions caused by the presence of the perspective of the frame. This, in turn, makes it possible to take photographs with the help of a camera located at an arbitrary angle to the reference and measured objects.
- the claimed method in contrast to the prototype, involves a fully automated measurement of the dimensions of an object of interest, for example, any part of the human body, due to the following steps: a) Automated detection of the contour of a human body part by analyzing all the contours in the photograph and selecting the largest contour in within the frame ; b) Automated detection of an accurate pixel mask of a human body part by processing the contour found above with an artificial neural network; c) Automated refinement of the contour due to the analysis of the found exact pixel mask of a part of the human body; d) Automated detection of measurement points on the contour of a human body part; e) Automated calculation of the actual dimensions of a human body part, taking into account perspective distortions of the frame caused by the location of the camera at an arbitrary angle to the measured object.
- human participation in calculating the size of a body part is limited to the stage of making photographs, and the subsequent stages: image analysis and calculations are carried out automatically by computer vision algorithms using an artificial neural network, which increases the accuracy of calculating the actual dimensions of the object being measured, reduces the processing time of photographs and excludes errors caused by the presence of the human factor.
- the method for measuring the linear size of the measured object according to the joint image of the measured and exemplary object consists of the following steps:
- the result of the algorithm execution is one size of the measured object (length or width) for one original image.
- the invention is illustrated by the following figures and an implementation example.
- the figure 1 shows an example of the location in the frame (in the original image) of the exemplary object 1 (for example, a sheet of A4 paper) and the measured object 2 (for example, a human foot) for the “top view” angle.
- the exemplary object 1 for example, a sheet of A4 paper
- the measured object 2 for example, a human foot
- the figure 2 shows an example of the original image of the reference object 1 and the measured object 2 for the side view.
- FIG. 1 An example of the implementation of the method.
- a mobile phone with a camera several photographs of the dorsum of the foot from above (Fig. 1) and from the side (Fig. 2) were sequentially taken.
- a model object 1 was placed, which was used as a sheet of A4 paper (the dimensions of which are 29.7 by 21.0 cm.).
- the resulting images were subjected to sequential automated processing in accordance with the claimed method, using computer vision algorithms and an artificial neural network. During processing, the mask and contour of the sheet of paper were automatically found, the contour was further approximated by a rectangle, for which the corner points were obtained.
- the mask and the contour of the foot were found, with the help of computer vision algorithms, the extreme points of the contour were determined (for top and side views).
- a projective transformation of the extreme points of the leaf and foot was constructed from the image plane, the distance between the projected points was measured, and the real physical dimensions of the human foot were obtained: length and width (27.2 and 10.0 cm, respectively).
- the resulting length and width of the foot can be converted into standard shoe sizes, greatly facilitating its selection.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Textile Engineering (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Description
Claims
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP22858840.6A EP4388912A4 (en) | 2021-08-19 | 2022-08-17 | METHOD FOR MEASURING THE LINEAR DIMENSIONS OF AN OBJECT FROM AN IMAGE |
| CN202280056626.7A CN117858641A (zh) | 2021-08-19 | 2022-08-17 | 基于图像测量对象的线性尺寸的方法 |
| US18/683,007 US20250139805A1 (en) | 2021-08-19 | 2022-08-17 | Method of Measuring The Linear Dimensions Of An Object On the Basis Of An Image |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| RU2021124620 | 2021-08-19 | ||
| RU2021124620A RU2777785C1 (ru) | 2021-08-19 | Способ измерения линейных размеров объекта по изображению с использованием образцового объекта |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023022630A1 true WO2023022630A1 (ru) | 2023-02-23 |
Family
ID=85240918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/RU2022/050252 Ceased WO2023022630A1 (ru) | 2021-08-19 | 2022-08-17 | Способ измерения линейных размеров объекта по изображению |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20250139805A1 (ru) |
| EP (1) | EP4388912A4 (ru) |
| CN (1) | CN117858641A (ru) |
| WO (1) | WO2023022630A1 (ru) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116452498A (zh) * | 2023-03-09 | 2023-07-18 | 内蒙古农业大学 | 一种改进U2-Net的显著性蒙古族图案检测算法 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080166019A1 (en) * | 2007-01-04 | 2008-07-10 | Lee Kual-Zheng | Systems and methods for object dimension estimation |
| US20140270540A1 (en) * | 2013-03-13 | 2014-09-18 | Mecommerce, Inc. | Determining dimension of target object in an image using reference object |
| JP2017101356A (ja) | 2015-12-03 | 2017-06-08 | フレックスジャパン株式会社 | 採寸方法及び採寸システム |
| US20190295147A1 (en) * | 2013-12-27 | 2019-09-26 | Ebay Inc. | Regional item recommendations |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102012204537A1 (de) * | 2012-03-21 | 2013-09-26 | rocket eleven GmbH | Verfahren zum Bestimmen der Abmessungen eines Körperteils |
| US20180247426A1 (en) * | 2015-08-28 | 2018-08-30 | Lawrence Gluck | System for accurate remote measurement |
| US20180160777A1 (en) * | 2016-12-14 | 2018-06-14 | Black Brass, Inc. | Foot measuring and sizing application |
-
2022
- 2022-08-17 CN CN202280056626.7A patent/CN117858641A/zh active Pending
- 2022-08-17 EP EP22858840.6A patent/EP4388912A4/en active Pending
- 2022-08-17 WO PCT/RU2022/050252 patent/WO2023022630A1/ru not_active Ceased
- 2022-08-17 US US18/683,007 patent/US20250139805A1/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080166019A1 (en) * | 2007-01-04 | 2008-07-10 | Lee Kual-Zheng | Systems and methods for object dimension estimation |
| US20140270540A1 (en) * | 2013-03-13 | 2014-09-18 | Mecommerce, Inc. | Determining dimension of target object in an image using reference object |
| US20190295147A1 (en) * | 2013-12-27 | 2019-09-26 | Ebay Inc. | Regional item recommendations |
| JP2017101356A (ja) | 2015-12-03 | 2017-06-08 | フレックスジャパン株式会社 | 採寸方法及び採寸システム |
Non-Patent Citations (2)
| Title |
|---|
| ROSEBROCK ADRIAN: "Measuring size of objects in an image with OpenCV - PyImageSearch", 28 March 2016 (2016-03-28), pages 1 - 149, XP055816807, Retrieved from the Internet <URL:https://www.pyimagesearch.com/2016/03/28/measuring-size-of-objects-in-an-image-with-opencv/> [retrieved on 20210622] * |
| See also references of EP4388912A4 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116452498A (zh) * | 2023-03-09 | 2023-07-18 | 内蒙古农业大学 | 一种改进U2-Net的显著性蒙古族图案检测算法 |
Also Published As
| Publication number | Publication date |
|---|---|
| EP4388912A1 (en) | 2024-06-26 |
| US20250139805A1 (en) | 2025-05-01 |
| CN117858641A (zh) | 2024-04-09 |
| EP4388912A4 (en) | 2025-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111649690B (zh) | 一种能够手持的3d信息采集的设备及方法 | |
| US10825198B2 (en) | 3 dimensional coordinates calculating apparatus, 3 dimensional coordinates calculating method, 3 dimensional distance measuring apparatus and 3 dimensional distance measuring method using images | |
| JP3624353B2 (ja) | 3次元形状計測方法およびその装置 | |
| CN107273846B (zh) | 一种人体体型参数确定方法及装置 | |
| US10424078B2 (en) | Height measuring system and method | |
| US10506225B2 (en) | Method of calibrating a camera | |
| CN103535960B (zh) | 基于数码图像的人体三维测量方法 | |
| WO2021185220A1 (zh) | 一种基于坐标测量的三维模型构建及测量方法 | |
| CN111028341B (zh) | 一种三维模型生成方法 | |
| CN102657531B (zh) | 基于计算机视觉的人体躯干围度测量方法和装置 | |
| JP2012042396A (ja) | 位置姿勢計測装置、位置姿勢計測方法、およびプログラム | |
| CN110986768B (zh) | 一种用于目标物3d信息高速采集测量设备 | |
| US12367600B2 (en) | Methods and systems for representing a user | |
| CN105378794A (zh) | 3d拍摄装置、用于建立3d图像的方法和用于构建3d拍摄装置的方法 | |
| CN107865473B (zh) | 人体特征测距装置及其测距方法 | |
| CN111127556A (zh) | 基于3d视觉的目标物体识别和位姿估算方法以及装置 | |
| CN111637834A (zh) | 一种三维数据测量装置及方法 | |
| CN111340959B (zh) | 一种基于直方图匹配的三维模型无缝纹理贴图方法 | |
| CN118642121A (zh) | 单目视觉测距与激光点云融合的空间定位方法及系统 | |
| WO2023022630A1 (ru) | Способ измерения линейных размеров объекта по изображению | |
| CN112435080A (zh) | 一种基于人体三维信息的虚拟制衣设备 | |
| CN112001973A (zh) | 基于数字散斑相关的快速三维人头测量方法 | |
| RU2777785C1 (ru) | Способ измерения линейных размеров объекта по изображению с использованием образцового объекта | |
| EA046631B1 (ru) | Способ измерения линейных размеров объекта по изображению с использованием образцового объекта | |
| CN111325780B (zh) | 一种基于图像筛选的3d模型快速构建方法 |
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: 22858840 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202490121 Country of ref document: EA |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 18683007 Country of ref document: US |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202280056626.7 Country of ref document: CN |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2022858840 Country of ref document: EP |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 2022858840 Country of ref document: EP Effective date: 20240319 |
|
| WWP | Wipo information: published in national office |
Ref document number: 18683007 Country of ref document: US |