CN115471769A - Visual identification method for existing state of tool in tool cabinet - Google Patents
Visual identification method for existing state of tool in tool cabinet Download PDFInfo
- Publication number
- CN115471769A CN115471769A CN202210978802.4A CN202210978802A CN115471769A CN 115471769 A CN115471769 A CN 115471769A CN 202210978802 A CN202210978802 A CN 202210978802A CN 115471769 A CN115471769 A CN 115471769A
- Authority
- CN
- China
- Prior art keywords
- image
- tool
- identification method
- threshold
- calculating
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a visual identification method for tool existing states in a tool cabinet, which comprises the steps of obtaining images in a fisheye camera video; correcting the original image, and cutting the corrected original image to obtain an image to be recognized; reading the calibration position, and cutting the tool from the image to be identified; comparing the real-time state of the tool with the standard state by using two recognition algorithms, and integrating the results; and sending the display effect to a front-end interface according to the integrated identification result. The method replaces the traditional 'tool three-checking' work in a manual mode, avoids the inefficient mode of manually checking the tools, and can meet the lean management requirements of tools in the fields of aviation, aerospace, ships, railways, electric power and the like.
Description
Technical Field
The application provides a visual identification method for tool existing states in a tool cabinet, and relates to the technical field of visual identification of tools.
Background
The tool is an important and indispensable asset for operators in any industry, and workers make daily contact with the tool, so that the tool is very important for enterprise management on how to quickly and conveniently apply, return, search and maintain the tool.
At present, tools are checked in a manual three-checking mode, the tools are various in types and large in quantity, checking work tasks are heavy, relevant responsible persons cannot be found after the tools are lost, and inconvenience and hidden dangers are brought to tool management.
Disclosure of Invention
The technical problem that this application will be solved is how to the automatic discernment of the existence state of instrument in the tool cabinet thereby the management to the instrument of assistance.
In order to solve the technical problems, the technical scheme of the application is to provide a visual identification method for the existing state of the tools in the tool cabinet, the pictures shot by a fisheye camera are processed, the tools are automatically identified, and then the identification result is displayed on a front-end interface, so that the efficient management of the tools is realized; the method specifically comprises the following steps:
a visual identification method for tool existence state in a tool cabinet is characterized by comprising the following steps:
the method comprises the following steps that firstly, a fisheye camera is used for obtaining a video stream, and an original image is extracted from the video stream;
performing circular detection on the original image, identifying an effective area of the fisheye camera, cutting the image according to the minimum circumscribed rectangle of the circular area, correcting the cut image, and unfolding the fisheye image into a normal view image for identifying the existence state of a subsequent tool;
thirdly, segmenting the corrected image according to the tool position corresponding to the pre-marked lining, segmenting the image at the single tool position from the image to obtain a single tool image, and taking the single tool image as an input image of the recognition algorithm;
comparing the separated single tool image with a standard reference image under the condition that the tool exists by respectively using an image chromatic aberration identification method and an image Hash identification method, obtaining a calibration threshold value before comparison, and calculating the two results through a combined evaluation formula to obtain an identification result;
and step five, according to the identification result, providing the identification information to the front end through the back end service for displaying to the user.
The calibration threshold in the fourth step is as follows:
all tools are taken out; calculating the calculated values under the two methods, and repeating the calculation for ten times;
all the tools are put back, the calculated values under the two methods are calculated, and the operation is repeated for ten times;
through the above operations, a plurality of values of the presence or absence of the tool are obtained, these values are placed on the axis, and the threshold value of the final tool state recognition is determined by taking the average of the presence and absence of the two extreme values: the threshold values of the image color difference identification method and the image hash identification method are respectively recorded as threshold 1 And threshold 2 With a nominal threshold value of alpha 1 threshold 1 +α 2 threshold 2 In which α is 1 ,α 2 Respectively, the combining weights.
Specifically, the formula for comparing the two images by the image color difference identification method is as follows, and R, G and B channels of the two images P1 and P2 are respectively taken out, wherein
Is the average of all elemental sums of the two image R channels,
refers to the average of all elemental sums of the R channel of P1,
in the same way;
the discrete cosine transform formula in the image Hash identification method calculation process is as follows, the image size is M multiplied by N, i, j is the row-column index of the image before discrete cosine transform, u, v is the row-column index of the image after transform, and f (i, j) is the pixel value of i row and j column:
calculating t2 from the transformed image F (u, v): reducing the size of the image F (u, v) to 32 × 32- > graying- > calculating DCT- > taking the upper left corner 8 × 8- > calculating the mean value- > comparing the DCT value with the mean value to obtain 0 and 1 matrix- > calculating the Hamming distance t2;
the combined evaluation formula is alpha 1 d(P 1 ,P 2 )+α 2 (1-t 2), comparison of α 1 d(P 1 ,P 2 )+α 2 (1-t 2) and a calibration threshold value, if alpha 1 d(P 1 ,P 2 )+α 2 And (1-t 2) judging that the tool exists if the value is larger than the calibration threshold value, and judging that the tool does not exist if the value is smaller than the calibration threshold value.
The method has the advantages that the threshold used for identifying the existing state of the tool is calibrated in advance, the influence caused by uneven illumination is avoided, specifically, a certain layer of tool to be identified is completely taken down or placed completely, the image chromatic aberration and the image hash under the two states are recorded, and the average value of the two states is calculated respectively to be used as the threshold for judging whether the tool exists or not. Then, the existing states of the tools are identified and the intersection is taken by using two identification algorithms respectively. The method and the device ensure that the identification result is more accurate by adopting the calibration threshold value, and greatly reduce the misjudgment rate.
The visual identification technology of the tool provided by the application is more intelligent and automatic for users, and the tool and the checking tool are automatically identified through an algorithm. Meanwhile, displaying the recognition result to a front-end interface; the records of the tool taken out and returned by the user are stored in the database, so that the user of the tool can be automatically tracked, and the complicated work of paper record is avoided.
Drawings
FIG. 1 is a flow chart of a method for visually identifying the presence of a tool in a tool cabinet according to the present application;
FIG. 2 is a schematic diagram illustrating identification of a fisheye camera in an original drawing;
FIG. 3 is a schematic diagram of the segmentation process for the corrected image corresponding to the tool position.
Detailed Description
In order to make the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Examples
The present embodiment provides a visual identification method for tool existing status in a tool cabinet, referring to fig. 1, the steps are briefly summarized as follows:
acquiring an image in a fisheye camera video;
correcting the original image, and cutting the corrected original image to obtain an image to be recognized;
reading the calibration position, and cutting the tool from the image to be identified;
comparing the real-time state of the tool with the standard state by using two recognition algorithms, and integrating the results;
and sending the display effect to a front-end interface according to the integrated recognition result.
The method comprises the steps of cutting and correcting an original image, separating tools from a picture according to the positions of all the tools calibrated in advance after the original image is obtained, specifically calculating the minimum enveloping circle of a fisheye image from the original image, cutting the circle, correcting the fisheye image into a common view angle image by using a correction algorithm after cutting, cutting off the part without the tools after correction, and taking the cut image as an image to be recognized.
As shown in fig. 2, the dotted line is an effective region photographed by the fisheye camera, and the thick line is a minimum circumscribed rectangular region.
The corrected image is subjected to tool part segmentation by pre-calibrated positions, as shown in fig. 3.
Specifically, the detailed steps of the visual recognition of the tool existing state are as follows:
s1, acquiring a video stream through a network video stream protocol by using a fisheye camera, and extracting an original image from the video stream.
S2, performing circle detection on the original image, identifying an effective area of the fisheye camera, cutting the image according to the minimum circumscribed rectangle of the circular area, correcting the cut image, and unfolding the fisheye image into a normal view image for identifying the existence state of a subsequent tool.
And S3, segmenting the corrected image according to the tool position corresponding to the pre-marked lining, segmenting the image at the single tool position from the image to obtain a single tool image, and taking the single tool image as an input image of the recognition algorithm.
And S4, respectively comparing the separated single tool image with a standard reference image in the presence of a tool (obtaining a calibration threshold value before comparison) by two methods (an image color difference identification method and an image hash identification method), and calculating the two results by a combined evaluation formula to obtain an identification result.
And S5, according to the identification result, providing the identification information to the front end through the back-end service for displaying to the user.
The method for calibrating the threshold in the step S4 is specifically as follows: the tool cabinet generally has a plurality of layers, and an independent fisheye camera is arranged in each layer, and because tools in each layer are different, for simplification, only the first layer is described in detail, and other layers are the same. All tools in the first layer are taken out, a request is sent to the back-end service, the back-end calculates the calculated values under the two algorithms, the operation is repeated for ten times, all tools in the first layer are put back, the request is sent to the back-end service, and the calculated values under the two algorithms are calculated by the back-end and returned for ten times. Through the twoIn the secondary operation, a plurality of numerical values of the existence of the tool can be obtained, the numerical values are placed on a numerical axis, the average value of the existence and the nonexistence of two extreme values is taken as the threshold value of the final tool state identification, and the threshold values of the image color difference identification method and the image hash identification method are respectively marked as threshold 1 And threshold 2 Then the final calibration threshold is: alpha is alpha 1 threshold 1 +α 2 threshold 2 In which α is 1 ,α 2 Respectively, the combining weights. The reason for calibrating the threshold value is that the manually set threshold value cannot be applied to different tools due to the influence of non-uniformity of a light source and a highly reflective tool, so that the tool needs to be subjected to self-adaptive threshold value calibration, and the self-adaptive threshold value calibration can greatly reduce the misjudgment rate.
Specifically, the formula for comparing two images by the image color difference recognition method is as follows, and R, G and B channels of two images P1 and P2 are respectively taken out, wherein
Is the average of all elemental sums of the two image R channels,
refers to the average of all elemental sums of the R channel of P1,
similarly.
The discrete cosine transform formula in the calculation process of the image hash identification method is as follows, wherein the image size is M multiplied by N, i, j is the row-column index of the image before discrete cosine transform, u, v is the row-column index of the image after the transform, and f (i, j) is the pixel value of i row, j column:
the combination evaluation formula is as follows
res＝α 1 threshold 1 +α 2 threshold 2 ＝α 1 d(P 1 ,P 2 )+α 2 (1-t2)
Wherein alpha is 1 ,α 2 Default to 0.5,threshold 1 And threshold 2 The threshold value of the image color difference identification method and the threshold value of the image hash identification method are respectively, wherein t2 is obtained in the following way: the transformed image F (u, v) obtained by the image Hash identification method is reduced to 32 × 32->Graying->Formula calculation DCT (discrete cosine transform) ->Get the upper left corner 8 x 8->Calculating the mean value->Comparing the DCT values with the mean value to obtain the 0 and 1 matrices>Calculating a Hamming distance t2;
if the calibration threshold is 0.65, assuming that d (P1, P2) is 0.7 and t2 is 0.8, the res calculation result is 0.45 and less than 0.65, and the tool is determined to be present, assuming that d (P1, P2) is 0.9 and t2 is 0.4, the res calculation result is 0.7 and more than 0.65, and the tool is determined to be absent.
Claims (3)
1. A visual identification method for tool existence state in a tool cabinet is characterized by comprising the following steps:
step one, acquiring a video stream by using a fisheye camera, and extracting an original image from the video stream;
performing circular detection on the original image, identifying an effective area of the fisheye camera, cutting the image according to the minimum circumscribed rectangle of the circular area, correcting the cut image, and unfolding the fisheye image into a normal view image for identifying the existence state of a subsequent tool;
thirdly, segmenting the corrected image according to the tool position corresponding to the pre-marked lining, segmenting the image at the single tool position from the image to obtain a single tool image, and taking the single tool image as an input image of the recognition algorithm;
comparing the separated single tool image with a standard reference image under the condition that the tool exists by respectively using an image chromatic aberration identification method and an image Hash identification method, obtaining a calibration threshold value before comparison, and calculating the two results through a combined evaluation formula to obtain an identification result;
and step five, according to the identification result, providing the identification information to the front end through the back-end service for displaying to the user.
2. The method according to claim 1, wherein the calibration thresholds in the fourth step are as follows:
all tools are taken out; calculating the calculated values under the two methods, and repeating for ten times;
all tools are put back, the calculated values under the two methods are calculated, and the operation is repeated for ten times;
by the above operations, a plurality of values of the presence or absence of the tool are obtained, these values are placed on the axes, and the threshold value of the final tool state recognition is determined by taking the average of the presence and absence of the two extreme values: the threshold values of the image color difference identification method and the image Hash identification method are respectively recorded as threshold 1 And threshold 2 With a nominal threshold value of alpha 1 threshold 1 +α 2 threshold 2 In which α is 1 ,α 2 Respectively, combining weights.
3. The visual recognition method of tool existence status in tool cabinet according to claim 2, wherein the formula of the image color difference recognition method comparing two images is as follows, R, G, B channels of two images P1 and P2 are taken out respectively, wherein
Is the average of all elemental sums of the two image R channels,
refers to the average of all elemental sums of the R channel of P1,
in the same way;
the discrete cosine transform formula in the image Hash identification method calculation process is as follows, the image size is M multiplied by N, i, j is the row-column index of the image before discrete cosine transform, u, v is the row-column index of the image after transform, and f (i, j) is the pixel value of i row and j column:
calculating t2 from the transformed image F (u, v): reducing the size of the image F (u, v) to 32 × 32- > graying- > calculating DCT- > taking the upper left corner 8 × 8- > calculating the mean value- > comparing the DCT value with the mean value to obtain 0 and 1 matrix- > calculating the Hamming distance t2;
the combined evaluation formula is alpha 1 d(P 1 ，P 2 )+α 2 (1-t 2), comparison of α 1 d(P 1 ,P 2 )+α 2 (1-t 2) and a calibration threshold value, if alpha 1 d(P 1 ,P 2 )+α 2 And (1-t 2) judging that the tool exists if the value is larger than the calibration threshold value, and judging that the tool does not exist if the value is smaller than the calibration threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210978802.4A CN115471769B (en) | 2022-08-16 | 2022-08-16 | Visual identification method for existing state of tool in tool cabinet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210978802.4A CN115471769B (en) | 2022-08-16 | 2022-08-16 | Visual identification method for existing state of tool in tool cabinet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115471769A true CN115471769A (en) | 2022-12-13 |
| CN115471769B CN115471769B (en) | 2023-04-07 |
Family
ID=84366488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210978802.4A Active CN115471769B (en) | 2022-08-16 | 2022-08-16 | Visual identification method for existing state of tool in tool cabinet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115471769B (en) |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101395592A (en) * | 2006-03-03 | 2009-03-25 | 美国唯美安视国际有限公司 | Movie studio-based network distribution system and method |
| CN103927804A (en) * | 2014-04-23 | 2014-07-16 | 国家电网公司 | Safe tool and instrument cabinet system |
| CN105141909A (en) * | 2015-08-17 | 2015-12-09 | 安徽四创电子股份有限公司 | Portal mobile image investigation device |
| US20180129865A1 (en) * | 2016-11-08 | 2018-05-10 | Nec Laboratories America, Inc. | Action recognition system with landmark localization on objects in images using convolutional neural networks |
| US20180232577A1 (en) * | 2017-02-13 | 2018-08-16 | Snap-On Incorporated | Automated tool data generation in automated asset management systems |
| JP2019104096A (en) * | 2017-12-14 | 2019-06-27 | 株式会社レイテクト | Tool storage shed |
| US20190213534A1 (en) * | 2018-01-10 | 2019-07-11 | Trax Technology Solutions Pte Ltd. | Withholding notifications due to temporary misplaced products |
| US20190311214A1 (en) * | 2018-04-05 | 2019-10-10 | Imagination Technologies Limited | Matching Local Image Feature Descriptors in Image Analysis |
| CN111126469A (en) * | 2019-12-17 | 2020-05-08 | 合肥鼎盛锦业科技有限公司 | Identification method and identification device of downhole tool and electronic equipment |
| US10719806B1 (en) * | 2019-05-07 | 2020-07-21 | RNB Controls, Inc. | Self-contained, transportable and reusable system, apparatus and method for automated securing and managing of tools at job sites |
| CN112005244A (en) * | 2019-11-15 | 2020-11-27 | 深圳市微蓝智能科技有限公司 | Article management method, terminal device, article management apparatus, and storage medium |
| CN112307786A (en) * | 2020-10-13 | 2021-02-02 | 上海迅邦电子科技有限公司 | Batch positioning and identifying method for multiple irregular two-dimensional codes |
| WO2021142388A1 (en) * | 2020-01-10 | 2021-07-15 | Adroit Worldwide Media, Inc. | System and methods for inventory management |
| CN113610480A (en) * | 2021-09-13 | 2021-11-05 | 中国南方电网有限责任公司超高压输电公司广州局 | Storage management system and method for electric power tools |
| US11199853B1 (en) * | 2018-07-11 | 2021-12-14 | AI Incorporated | Versatile mobile platform |
| CN113920461A (en) * | 2021-10-09 | 2022-01-11 | 国网甘肃省电力公司电力科学研究院 | Power grid operation and maintenance process image monitoring system and monitoring method |
| CN113974837A (en) * | 2021-11-02 | 2022-01-28 | 武汉联影智融医疗科技有限公司 | Identification system and method of end tool, end tool and surgical robot system |
| CN114202858A (en) * | 2021-11-23 | 2022-03-18 | 东莞职业技术学院 | Tool management cabinet and control method thereof |
-
2022
- 2022-08-16 CN CN202210978802.4A patent/CN115471769B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101395592A (en) * | 2006-03-03 | 2009-03-25 | 美国唯美安视国际有限公司 | Movie studio-based network distribution system and method |
| CN103927804A (en) * | 2014-04-23 | 2014-07-16 | 国家电网公司 | Safe tool and instrument cabinet system |
| CN105141909A (en) * | 2015-08-17 | 2015-12-09 | 安徽四创电子股份有限公司 | Portal mobile image investigation device |
| US20180129865A1 (en) * | 2016-11-08 | 2018-05-10 | Nec Laboratories America, Inc. | Action recognition system with landmark localization on objects in images using convolutional neural networks |
| US20180232577A1 (en) * | 2017-02-13 | 2018-08-16 | Snap-On Incorporated | Automated tool data generation in automated asset management systems |
| JP2019104096A (en) * | 2017-12-14 | 2019-06-27 | 株式会社レイテクト | Tool storage shed |
| US20190213534A1 (en) * | 2018-01-10 | 2019-07-11 | Trax Technology Solutions Pte Ltd. | Withholding notifications due to temporary misplaced products |
| US20190311214A1 (en) * | 2018-04-05 | 2019-10-10 | Imagination Technologies Limited | Matching Local Image Feature Descriptors in Image Analysis |
| US11199853B1 (en) * | 2018-07-11 | 2021-12-14 | AI Incorporated | Versatile mobile platform |
| US10719806B1 (en) * | 2019-05-07 | 2020-07-21 | RNB Controls, Inc. | Self-contained, transportable and reusable system, apparatus and method for automated securing and managing of tools at job sites |
| WO2021092883A1 (en) * | 2019-11-15 | 2021-05-20 | 深圳市微蓝智能科技有限公司 | Article management method, terminal apparatus, article management device, and storage medium |
| CN112005244A (en) * | 2019-11-15 | 2020-11-27 | 深圳市微蓝智能科技有限公司 | Article management method, terminal device, article management apparatus, and storage medium |
| CN111126469A (en) * | 2019-12-17 | 2020-05-08 | 合肥鼎盛锦业科技有限公司 | Identification method and identification device of downhole tool and electronic equipment |
| WO2021142388A1 (en) * | 2020-01-10 | 2021-07-15 | Adroit Worldwide Media, Inc. | System and methods for inventory management |
| CN112307786A (en) * | 2020-10-13 | 2021-02-02 | 上海迅邦电子科技有限公司 | Batch positioning and identifying method for multiple irregular two-dimensional codes |
| CN113610480A (en) * | 2021-09-13 | 2021-11-05 | 中国南方电网有限责任公司超高压输电公司广州局 | Storage management system and method for electric power tools |
| CN113920461A (en) * | 2021-10-09 | 2022-01-11 | 国网甘肃省电力公司电力科学研究院 | Power grid operation and maintenance process image monitoring system and monitoring method |
| CN113974837A (en) * | 2021-11-02 | 2022-01-28 | 武汉联影智融医疗科技有限公司 | Identification system and method of end tool, end tool and surgical robot system |
| CN114202858A (en) * | 2021-11-23 | 2022-03-18 | 东莞职业技术学院 | Tool management cabinet and control method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115471769B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764071B (en) | Real face detection method and device based on infrared and visible light images | |
| US9177194B2 (en) | System and method for visually distinguishing faces in a digital image | |
| US7657090B2 (en) | Region detecting method and region detecting apparatus | |
| US7643686B2 (en) | Multi-tiered image clustering by event | |
| US6246790B1 (en) | Image indexing using color correlograms | |
| US7676085B2 (en) | Method and apparatus for representing a group of images | |
| CN110267061B (en) | News splitting method and system | |
| EP1300804A2 (en) | Face detecting method by skin color recognition | |
| US20060098256A1 (en) | Method of and system for image processing and computer program | |
| EP1139288A2 (en) | A method of detecting duplicate pictures in an automatic albuming system | |
| US20110001850A1 (en) | Automatic Redeye Detection | |
| US20060204082A1 (en) | Fusion of color space data to extract dominant color | |
| US5642443A (en) | Whole order orientation method and apparatus | |
| US6411339B1 (en) | Method of spatio-temporally integrating/managing a plurality of videos and system for embodying the same, and recording medium for recording a program for the method | |
| US20190311744A1 (en) | Comparing frame data to generate a textless version of a multimedia production | |
| US6882746B1 (en) | Normalized bitmap representation of visual object's shape for search/query/filtering applications | |
| CN115471769B (en) | Visual identification method for existing state of tool in tool cabinet | |
| CN105912997A (en) | Face recognition method and system | |
| CN111708907B (en) | Target person query method, device, equipment and storage medium | |
| JPWO2018173947A1 (en) | Image retrieval device | |
| US20040012576A1 (en) | Digital image display method and system | |
| US20210295056A1 (en) | Method, apparatus, electronic device, and computer storage medium for video processing | |
| CN113178079B (en) | Marking system, method and storage medium for signal lamp and lane line | |
| CN111586363B (en) | Video file viewing method and system based on object | |
| CN110266939B (en) | Display method, electronic device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |