WO2011151477A1 - Procedimiento de reconocimiento de objetos - Google Patents
Procedimiento de reconocimiento de objetos Download PDFInfo
- Publication number
- WO2011151477A1 WO2011151477A1 PCT/ES2010/070369 ES2010070369W WO2011151477A1 WO 2011151477 A1 WO2011151477 A1 WO 2011151477A1 ES 2010070369 W ES2010070369 W ES 2010070369W WO 2011151477 A1 WO2011151477 A1 WO 2011151477A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- module
- storage module
- stored
- objects
- digital
- 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/28—Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/772—Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
Definitions
- the present invention relates, in general, to a method of automatic recognition and location of objects within a storage module.
- This procedure includes the development of a set of patterns, one pattern per object to be identified that are stored in a storage memory or database.
- Each pattern includes essential characters of an objective object of recognition and location.
- the array of captured data images is compared with each pattern corresponding to the set of target objects to be recognized and located in the shelf module.
- the procedure finds a match between the captured image and some pattern, the procedure generates an output signal indicating the object / s recognized in the shelf module.
- a disadvantage of the above recognition procedure is to recognize the position and orientation of an objective object where the object can be one of a set of objective objects.
- the present invention seeks to solve or reduce one or more of the drawbacks set forth above by a method of recognizing and locating objects within a storage module as defined in the claims.
- the recognition and location procedure identifies from a digital image captured by a digital camera the position and orientation of a target object stored in a variety of positions, without digitally storing a wide variety of essential characters by pattern associated with the objective object.
- the procedure recognizes and locates an objective object in known position and orientation with respect to the camera seeking the match between the digital data matrix associated with the digital image captured with a pattern comprising a small number of characters or Essential features in a position and orientation.
- the object may be rotated with respect to the stored pattern since an objective is not an objective of the recognition procedure to determine the orientation of the object, but to recognize objective objects and their location within the shelf module.
- the search for coincidence within a reduced pattern can be performed in a shorter time and, consequently, the calculation time required to compare the data matrix with the object object patterns is significantly reduced.
- an operator or point-of-sale manager captures digital images of objective objects stored in a storage shelf module to recognize and locate within it.
- the manager performs image capture by means of image capture such as a digital camera, for example.
- the manager is located in front of the module under study and at a predetermined distance from it. That is, the image capture is carried out frontally to the shelf module without using a zoom and / or an additional light source internal and / or external to the digital camera, that is, with the ambient light existing in the module location shelves
- the camera must be able to capture digital images with a resolution around 5 Megapixels, therefore, a digital image is a matrix of pixels.
- sequence of digital images must be captured according to a predetermined sequence to subsequently generate a single image matrix of digital data associated with all the digital images corresponding to the module of shelves studied.
- the image capture medium comprises a general purpose processing module such as a microprocessor programmed to verify whether the captured images meet a set of technical characteristics or minimum quality requirements such as predetermined focus and brightness.
- a general purpose processing module such as a microprocessor programmed to verify whether the captured images meet a set of technical characteristics or minimum quality requirements such as predetermined focus and brightness.
- the microprocessor runs a stored program to verify if each of the captured images meets the minimum quality requirements. If the verification is positive, the microprocessor generates a first output message to indicate the suitability of the captured images, if not, it generates a second message to indicate the suitability of capturing a second image corresponding to the shelf module or part thereof whose First image is inappropriate for the next steps of the automatic recognition and localization process.
- all digital images captured and stored are transmitted from the capture module through a first module input / output to a first application module or application server via an air interface associated with a wireless or wireless telecommunications system, for example.
- the server is connected, in turn, to a computer unit comprising a general purpose computer programmed to perform the recognition analysis of the stored images.
- the computer comprises a second central processing unit, a multimedia screen, a keyboard for the input of images stored in the first memory.
- the second processor is programmed to perform the recognition analysis of the captured images.
- the first step performed by the second processor is to generate a single array of digital data comprising elements of the array associated with the pixels of the images captured per module of shelves under study.
- the second processor executes a subroutine to identify technical characteristics associated with common points existing in each matrix of pixels captured by the point of sale agent and associated to the module under study.
- the captured images must be captured according to certain image capture criteria so that at least one point in common can be established between sequentially captured images.
- Said point in common can be, for example, a particular edge of a shelf corresponding to the shelf module, so that, if in an image, said edge is to the right of the captured image, in the adjacent captured image The same border will be to the left of the next captured image.
- the common point may also be an object stored on a shelf in a known position and orientation with respect to the camera.
- the global pixel array generated from a set of image matrices comprises a set of objects in a known position and orientation with respect to the camera, from which to recognize and locate target objects. within said shelf module, by executing a predetermined recognition and location guideline.
- the second processor executes a recognition procedure comprising the selection from a database of a subset of stored patterns associated with objects to be recognized and located within the shelf module, so that a certain stored pattern is compared with a subset of elements of the digital data matrix generated in the previous stage of the procedure. The result of the comparison is to identify the existence of coincidences between a subset of matrix elements associated with an objective object and a stored pattern and associated with an objective object.
- Each pattern stored in the first memory comprises a set of differential or invariant characteristics obtained from the digital image of a target object.
- the stored pattern is generated from the image captured in a controlled environment.
- the pixel matrices have been transformed into an array of data comprising essential characters or features in a position and orientation with respect to the camera of each of the objective objects stored in the shelves.
- each stored pattern associated with an object comprises the essential characters that can be searched in the data matrix.
- the second processor executes the recognition and location directive, it can detect some coincidence between the set of elements of the global matrix associated with an objective object and a stored pattern relative to the same object.
- the second processor executes the recognition directive so as to minimize the elements of the matrix that are subjected to comparison with the stored patterns, that is, it performs a process of conditioning the matrix generated from the images digital captured in the store, to be able to discard those areas within the storage module where the second processor determines the absence of products. Therefore, the second processor discards those areas in which there is a very high regularity and, consequently, these elements of the matrix are associated with an area where there is no possibility that there is a product since no changes in the characteristics or details are detected , for example, a bottom wall of the module that is an all white surface or a very dark area.
- the objective objects are compared with the stored patterns.
- essential characteristics are extracted from the patterns that are compared with matrix elements associated with objects.
- the element is associated with an area of the module where an object can exist;
- the second processor performs a search for other essential characteristics of the pattern around the matrix element where a match has been detected.
- the recognition guideline includes a semantic recognition subroutine associated with the labels of the target objects.
- the second processor finds a minimum number of matches between the essential features of the target object and the stored pattern, it generates a third output message indicating the identified target object and its position within the shelf module.
- a pattern may comprise a number of essential characteristics that is a function of the type of object to which it is associated, that is, characteristics that can differentiate said object from a different object, such as characteristic parameters of the physical form and dimensions of the object. , characteristics associated with the labeling of the object, etc.
- the generation of the third output message showing a match is only generated if a minimum number of matches are recognized between the essential characteristics of a given pattern and an objective object. In the event that the number of matches is lower than the defined match threshold, it would mean that it is not possible to recognize a certain object.
- the second processor executes a subroutine of the recognition directive to find matches between essential characteristics of an object adjacent to the previous one with the stored pattern in order to recognize if the adjacent object comprises a number of essential characteristics greater than threshold of matches identified with the adjacent object and, in addition, if among the identified matches are the essential characteristics of the object that does not exceed the threshold of matches, the recognition guideline determines that both adjacent objects coincide with the same stored pattern.
- the second processor determines that one of the analyzed target objects is rotated with respect to the stored pattern and the adjacent object has no rotation with respect to the same pattern.
- the described scenario can be found, for example, in a linear of a large surface where several objects of the same product are stored and aligned in the same shelf, where a consumer has taken one of them and then returns the object to its initial position but rotated with respect to the initial position. However, the remaining adjacent objects have remained in their initial position corresponding to the position and orientation that contains the maximum number of matches with the pattern associated with that type of object.
- the previous step can be repeated several times while identifying several coincidences of essential characteristics between the analyzed object and the stored pattern, since a shelf of a linear of a large area stores several identical objects in the same shelf and on adjacent shelves in height.
- the second processor Once all the positions of the data matrix have been analyzed by the recognition directive executed by the second processor to recognize matches with the stored patterns associated with the objective objects; the second processor generates a number of messages relative to the number of matches detected; being able to generate from these messages a report with the results obtained, so that an image of the shelf module appears with the identified target objects and their position within it and the image of the shelf module with the desired objects and placed in the desired position.
- target objects of a child audience such as popcorn should be placed on lower shelves of the shelf module, for example.
- the diagram or planogram is a graphic representation of how a product category is exposed in a linear, gondola or display stand. With these planograms a subsequent analysis of the main variables that are produced at the point of sale such as assortment present in each store, percentage of space that is dedicated in an implantation to each product, number of faces per product, adjacencies of products, placement rules used, etc., compared to how an objective planogram of how products should be placed.
- planograms Once the planograms have been generated, they can be entered into a database for later analysis and report generation that will reflect on an individual level or aggregate the situation of the main variables that occur at a time in the point of sale.
- the procedure for recognizing objects stored in a storage module of a point of sale of products can be performed by a computer, loadable within an internal memory of a computer with input and output units and, also, with units of processors.
- the computer program comprises for this purpose codes configured to execute the steps of the above process when executed by the computer.
- executable codes can be recorded on a readable carrier medium within a computer.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
Description
Claims
Priority Applications (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/ES2010/070369 WO2011151477A1 (es) | 2010-06-01 | 2010-06-01 | Procedimiento de reconocimiento de objetos |
| JP2013512949A JP2013528869A (ja) | 2010-06-01 | 2010-06-01 | オブジェクトを認識する手続き |
| EP10852455.4A EP2579209A4 (en) | 2010-06-01 | 2010-06-01 | METHOD FOR RECOGNIZING OBJECTS |
| BR112012008408A BR112012008408A2 (pt) | 2010-06-01 | 2010-06-01 | método para reconhecer objetos, dispositivo legível por computador e programa de computador |
| US13/394,486 US20120170809A1 (en) | 2010-06-01 | 2010-06-01 | Procedure for recognizing objects |
| MX2012003092A MX337257B (es) | 2010-06-01 | 2010-06-01 | Procedimiento de reconocimiento de objetos. |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/ES2010/070369 WO2011151477A1 (es) | 2010-06-01 | 2010-06-01 | Procedimiento de reconocimiento de objetos |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2011151477A1 true WO2011151477A1 (es) | 2011-12-08 |
Family
ID=45066209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/ES2010/070369 Ceased WO2011151477A1 (es) | 2010-06-01 | 2010-06-01 | Procedimiento de reconocimiento de objetos |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US20120170809A1 (es) |
| EP (1) | EP2579209A4 (es) |
| JP (1) | JP2013528869A (es) |
| BR (1) | BR112012008408A2 (es) |
| MX (1) | MX337257B (es) |
| WO (1) | WO2011151477A1 (es) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12265976B1 (en) * | 2021-07-07 | 2025-04-01 | Blue Yonder Group, Inc. | Computer vision in-store planogram analysis |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013212333A1 (de) * | 2013-06-26 | 2014-12-31 | Bayerische Motoren Werke Aktiengesellschaft | Verfahren und Vorrichtung zur Überwachung einer Teileentnahme, Teilebereitstellungssystem, Verwendung eines Vibrationsalarmgeräts |
| EP3848867A1 (en) | 2015-03-18 | 2021-07-14 | United Parcel Service Of America, Inc. | Systems and methods for verifying the contents of a shipment |
| US10366364B2 (en) | 2015-04-16 | 2019-07-30 | United Parcel Service Of America, Inc. | Enhanced multi-layer cargo screening system, computer program product, and method of using the same |
| US11109065B2 (en) * | 2018-09-26 | 2021-08-31 | Google Llc | Video encoding by providing geometric proxies |
| CN109919040B (zh) * | 2019-02-15 | 2022-04-19 | 北京清瞳时代科技有限公司 | 货物的品规信息识别方法及装置 |
| US10867190B1 (en) * | 2019-11-27 | 2020-12-15 | Aimotive Kft. | Method and system for lane detection |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010017945A1 (en) * | 2000-02-03 | 2001-08-30 | Minolta Co., Ltd. | Image processing device, image processing method and image processing program for joining a plurality of divided images |
| WO2007130688A2 (en) * | 2006-05-10 | 2007-11-15 | Evolution Robotics, Inc. | Mobile computing device with imaging capability |
| WO2008091281A2 (en) * | 2006-08-01 | 2008-07-31 | California Institute Of Technology | Artificial intelligence systems for identifying objects |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US2004A (en) * | 1841-03-12 | Improvement in the manner of constructing and propelling steam-vessels | ||
| US4414685A (en) * | 1979-09-10 | 1983-11-08 | Sternberg Stanley R | Method and apparatus for pattern recognition and detection |
| US7106885B2 (en) * | 2000-09-08 | 2006-09-12 | Carecord Technologies, Inc. | Method and apparatus for subject physical position and security determination |
| JP2007034872A (ja) * | 2005-07-29 | 2007-02-08 | Matsushita Electric Ind Co Ltd | 顔認証装置および顔認証システムならびに顔認証方法 |
| WO2007117368A2 (en) * | 2006-04-12 | 2007-10-18 | Store Eyes, Inc. | Method for measuring retail display and compliance |
| JP2009187482A (ja) * | 2008-02-08 | 2009-08-20 | Nippon Sogo System Kk | 棚割再現方法、棚割再現プログラム、棚割評価方法、棚割評価プログラム及び記録媒体 |
| RU2395787C2 (ru) * | 2008-05-19 | 2010-07-27 | Государственный Научный Центр России "Государственное Научное Учреждение "Центральный Научно-Исследовательский И Опытно-Конструкторский Институт Робототехники И Технической Кибернетики"(Цнии Ртк) | Способ обнаружения объектов |
| US9147174B2 (en) * | 2008-08-08 | 2015-09-29 | Snap-On Incorporated | Image-based inventory control system using advanced image recognition |
-
2010
- 2010-06-01 WO PCT/ES2010/070369 patent/WO2011151477A1/es not_active Ceased
- 2010-06-01 US US13/394,486 patent/US20120170809A1/en not_active Abandoned
- 2010-06-01 JP JP2013512949A patent/JP2013528869A/ja active Pending
- 2010-06-01 BR BR112012008408A patent/BR112012008408A2/pt not_active IP Right Cessation
- 2010-06-01 MX MX2012003092A patent/MX337257B/es active IP Right Grant
- 2010-06-01 EP EP10852455.4A patent/EP2579209A4/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010017945A1 (en) * | 2000-02-03 | 2001-08-30 | Minolta Co., Ltd. | Image processing device, image processing method and image processing program for joining a plurality of divided images |
| WO2007130688A2 (en) * | 2006-05-10 | 2007-11-15 | Evolution Robotics, Inc. | Mobile computing device with imaging capability |
| WO2008091281A2 (en) * | 2006-08-01 | 2008-07-31 | California Institute Of Technology | Artificial intelligence systems for identifying objects |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP2579209A4 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12265976B1 (en) * | 2021-07-07 | 2025-04-01 | Blue Yonder Group, Inc. | Computer vision in-store planogram analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| US20120170809A1 (en) | 2012-07-05 |
| EP2579209A4 (en) | 2014-01-15 |
| BR112012008408A2 (pt) | 2016-03-29 |
| JP2013528869A (ja) | 2013-07-11 |
| EP2579209A1 (en) | 2013-04-10 |
| MX337257B (es) | 2014-09-23 |
| MX2012003092A (es) | 2012-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2011151477A1 (es) | Procedimiento de reconocimiento de objetos | |
| US11900316B2 (en) | Information processing apparatus, control method, and program | |
| Xu et al. | Human re-identification by matching compositional template with cluster sampling | |
| US7986828B2 (en) | People detection in video and image data | |
| US9361520B2 (en) | Method and system for tracking objects | |
| CN104376575B (zh) | 一种基于多摄像头监控的行人计数方法和装置 | |
| WO2020186520A1 (zh) | 价签信息显示方法、装置及货架系统 | |
| CN114743026B (zh) | 目标对象的方位检测方法、装置、设备及计算机可读介质 | |
| US11775788B2 (en) | Arbitrary visual features as fiducial elements | |
| CN109416744A (zh) | 改进的相机校准系统、目标和过程 | |
| US11684846B2 (en) | Tracking playing cards during game play using RFID tags | |
| US11055539B2 (en) | Image processing for distinguishing individuals in groups | |
| CN110189343B (zh) | 图像标注方法、装置及系统 | |
| US20190213755A1 (en) | Image labeling for cleaning robot deep learning system | |
| CN112287867A (zh) | 一种多摄像头的人体动作识别方法及装置 | |
| CN112200851A (zh) | 一种基于点云的目标检测方法、装置及其电子设备 | |
| WO2016190783A1 (ru) | Способ визуализации объекта | |
| CN114764814B (zh) | 植物高度确定方法、装置、电子设备及介质 | |
| US20200402278A1 (en) | Flow line display system, flow line display method, and program recording medium | |
| CN112230765A (zh) | Ar显示方法、ar显示设备和计算机可读存储介质 | |
| KR102421604B1 (ko) | 이미지 처리 방법, 장치 및 전자 기기 | |
| CN117557901A (zh) | 一种田间小目标作物的检测模型及构建方法 | |
| Santos et al. | Multiple camera people detection and tracking using support integration | |
| JP6304815B2 (ja) | 画像処理装置ならびにその画像特徴検出方法、プログラムおよび装置 | |
| US20120038602A1 (en) | Advertisement display system and method |
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: 10852455 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2010852455 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 13394486 Country of ref document: US |
|
| WWE | Wipo information: entry into national phase |
Ref document number: MX/A/2012/003092 Country of ref document: MX |
|
| ENP | Entry into the national phase |
Ref document number: 2013512949 Country of ref document: JP Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112012008408 Country of ref document: BR |
|
| ENP | Entry into the national phase |
Ref document number: 112012008408 Country of ref document: BR Kind code of ref document: A2 Effective date: 20120410 |