CN113221808A - Dinner plate counting statistical method and device based on image recognition - Google Patents
Dinner plate counting statistical method and device based on image recognition Download PDFInfo
- Publication number
- CN113221808A CN113221808A CN202110575488.0A CN202110575488A CN113221808A CN 113221808 A CN113221808 A CN 113221808A CN 202110575488 A CN202110575488 A CN 202110575488A CN 113221808 A CN113221808 A CN 113221808A
- Authority
- CN
- China
- Prior art keywords
- dinner plate
- dinner
- algorithm
- tracking
- counting
- 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.)
- Withdrawn
Links
- 235000003166 Opuntia robusta Nutrition 0.000 title claims abstract description 67
- 244000218514 Opuntia robusta Species 0.000 title claims abstract description 67
- 238000007619 statistical method Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 10
- 238000001914 filtration Methods 0.000 claims abstract description 8
- 230000005540 biological transmission Effects 0.000 claims abstract description 3
- 235000012054 meals Nutrition 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 8
- 230000000903 blocking effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 206010070834 Sensitisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 230000008313 sensitization Effects 0.000 description 1
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
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a dinner plate counting statistical method and a dinner plate counting statistical device based on image recognition, wherein the method comprises the following steps: constructing a network model of dinner plate identification; acquiring a video image of a dinner plate on a transmission device, and reading a video frame; identifying dinner plates in each video frame by using a network model, tracking the identified dinner plates by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching lost frames by using a KFC algorithm; the dishes were counted using the electronic fence technique. The invention can obviously improve the accuracy of the detection and counting of the dinner plate.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a dinner plate counting statistical method and a dinner plate counting statistical device based on image recognition.
Background
The current system of counting statistics to the dinner plate is mostly the light sense counter, counts through placing light emitting component and light sense subassembly in conveyer belt both sides, can cover light emitting component's light when the dinner plate passes through the light sense subassembly, through the back, the light sense subassembly is sensitization once more, the count adds 1, accomplishes a count operation, but its count principle has certain limitation, just two dinner plates can not shelter from between light emitting component and light sense subassembly, otherwise the count can reduce, appear the count deviation.
Disclosure of Invention
The invention aims to provide a dinner plate counting statistical method and a dinner plate counting statistical device based on image recognition, and aims to overcome the defects in the prior art.
In order to achieve the technical purpose, the technical scheme of the invention is realized as follows:
a dinner plate counting statistical method based on image recognition, the method comprising the following steps:
constructing a network model of dinner plate identification;
acquiring a video image of a dinner plate on a transmission device, and reading a video frame;
identifying dinner plates in each video frame by using a network model, tracking the identified dinner plates by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching lost frames by using a KFC algorithm;
the dishes were counted using the electronic fence technique.
A meal tray count statistics apparatus based on image recognition, the apparatus comprising:
the dinner plate conveying module is used for conveying dinner plates;
the image acquisition module is used for acquiring a video image of the dinner plate and sending the video image to the dinner plate identification module;
the dinner plate identification module is used for reading the video frames and identifying dinner plates in the video frames;
the dinner plate tracking module is used for tracking the identified dinner plate by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching a lost frame by using a KFC algorithm;
and the dinner plate counting module is used for counting dinner plates by using the electronic fence technology.
The invention has the beneficial effects that: the invention can obviously improve the accuracy of the detection and counting of the dinner plate.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the invention;
FIG. 2 is a diagram illustrating an example of an application of the method according to an embodiment of the present invention;
fig. 3 is a block diagram of functional module structures of the apparatus according to the embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention is clearly and completely described below with reference to the drawings in the embodiments of the present invention.
As shown in fig. 1, according to an embodiment of the present invention, a meal tray counting statistical method based on image recognition includes the following steps:
and constructing a network model of the dinner plate identification.
And acquiring a video image of the dinner plate on the conveying device, and reading a video frame.
And identifying the dinner plate in each video frame by using a network model, tracking the identified dinner plate by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching the lost frame by using a KFC algorithm. Specifically, the dinner plate is identified by using a network model Yolov3, the result is in the form of a target position, a confidence degree and a classification confidence degree, the basic structure is a feature extraction layer based on a convolutional neural network and a target detection structure based on an anchor, the detection of the target position takes the fixed anchor as a reference, a deviation value is regressed to obtain a final result, and a plurality of different confidence degrees of the target need to be processed by a sigmoid function after being output.
The obtained target detection result of each frame is processed by adopting a tracking algorithm, specifically using a Kalman filtering algorithm, and the details are as follows:
prediction model
Here we describe the object model, i.e. the representation and the use for propagating the identification of the target to the next frame. The inter-frame displacement we approximate has a linear iso-velocity model independent of other objects and camera motion. The model for each target of the state is as follows:
where u and v represent the central horizontal and vertical coordinates of the object, respectively, and s and r represent the dimensional size and scale of the BBox of the object, note that the aspect ratio should be a constant. The latter three quantities thus represent the predicted next frame, and the detected bounding box is used to update the target state when detecting the association with the target, where the velocity component is solved for optimization by the kalman method. If no detection is associated with the target, only the linear velocity model need be used.
Data association
And performing data association by using a Hungarian assignment algorithm, wherein the used cost matrix is the IOU between the predicted position of the original target in the current frame and the target detection box of the current frame, and the assignment result which is smaller than the assigned IOU threshold value is invalid. The IOU is used for solving the problem that the target is blocked for a short time, because when the target is blocked, the blocking object is detected, the original target is not detected, and the blocking object and the original target are supposed to be associated. Then after occlusion ends, the correct association can be restored quickly because the similarly sized target IOU tends to be larger. This is based on the fact that the area of the shade is larger than the target.
If successive T frames do not achieve IOU matching of the tracked target predicted location and the detection box, the target is considered to be lost. T is set to be 1 in the system for two reasons, wherein the assumption of uniform motion is unreasonable, and the system mainly focuses on short-time target tracking. At this time, deleting the lost target as early as possible helps to improve the tracking efficiency. However, problems arise whereby the ID of the target must be switched frequently, which can cause inaccuracies in the trace count. At the moment, a KFC algorithm is used for matching lost frames, so that the accuracy of tracking and counting is ensured.
The dishes were counted using the electronic fence technique. Specifically, as shown in fig. 2, the dinner plate in the video moves horizontally, and due to the participation of the tracking algorithm, each box will obtain a digital ID identification, and each time one box crosses a set line (electronic fence), the count is incremented, and the ID is recorded.
As shown in fig. 3, based on the above dinner plate counting statistical method based on image recognition, the present invention also discloses a dinner plate counting statistical device based on image recognition, the device includes:
the dinner plate conveying module is used for conveying dinner plates;
the image acquisition module is used for acquiring a video image of the dinner plate and sending the video image to the dinner plate identification module;
the dinner plate identification module is used for reading the video frames and identifying dinner plates in the video frames;
the dinner plate tracking module is used for tracking the identified dinner plate by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching a lost frame by using a KFC algorithm;
and the dinner plate counting module is used for counting dinner plates by using the electronic fence technology.
The functional modules of the dinner plate counting and counting device based on image recognition in the embodiment of the invention respectively correspond to the operation steps of the dinner plate counting and counting method based on image recognition, and are not described again here.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (2)
1. A dinner plate counting statistical method based on image recognition is characterized by comprising the following steps:
1) constructing a network model of dinner plate identification;
2) acquiring a video image of a dinner plate on a transmission device, and reading a video frame;
3) identifying dinner plates in each video frame by using a network model, tracking the identified dinner plates by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching lost frames by using a KFC algorithm;
4) the dishes were counted using the electronic fence technique.
2. An image recognition-based meal tray count statistics apparatus, the apparatus comprising:
the dinner plate conveying module is used for conveying dinner plates;
the image acquisition module is used for acquiring a video image of the dinner plate and sending the video image to the dinner plate identification module;
the dinner plate identification module is used for reading the video frames and identifying dinner plates in the video frames;
the dinner plate tracking module is used for tracking the identified dinner plate by using a Kalman filtering tracking algorithm, performing data association by using a Hungarian algorithm and matching a lost frame by using a KFC algorithm;
and the dinner plate counting module is used for counting dinner plates by using the electronic fence technology.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110575488.0A CN113221808A (en) | 2021-05-26 | 2021-05-26 | Dinner plate counting statistical method and device based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110575488.0A CN113221808A (en) | 2021-05-26 | 2021-05-26 | Dinner plate counting statistical method and device based on image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113221808A true CN113221808A (en) | 2021-08-06 |
Family
ID=77098581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110575488.0A Withdrawn CN113221808A (en) | 2021-05-26 | 2021-05-26 | Dinner plate counting statistical method and device based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113221808A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835323A (en) * | 2015-05-19 | 2015-08-12 | 银江股份有限公司 | Multi-target public transport passenger flow detection method combining with electronic fence |
| CN109522854A (en) * | 2018-11-22 | 2019-03-26 | 广州众聚智能科技有限公司 | A kind of pedestrian traffic statistical method based on deep learning and multiple target tracking |
| CN110852283A (en) * | 2019-11-14 | 2020-02-28 | 南京工程学院 | Helmet wearing detection and tracking method based on improved YOLOv3 |
| CN111860282A (en) * | 2020-07-15 | 2020-10-30 | 中国电子科技集团公司第三十八研究所 | Subway section passenger flow volume statistics and pedestrian retrograde motion detection method and system |
| CN111986237A (en) * | 2020-09-01 | 2020-11-24 | 安徽炬视科技有限公司 | Real-time multi-target tracking algorithm irrelevant to number of people |
| CN112669349A (en) * | 2020-12-25 | 2021-04-16 | 北京竞业达数码科技股份有限公司 | Passenger flow statistical method, electronic equipment and storage medium |
| CN112836639A (en) * | 2021-02-03 | 2021-05-25 | 江南大学 | Pedestrian multi-target tracking video identification method based on improved YOLOv3 model |
-
2021
- 2021-05-26 CN CN202110575488.0A patent/CN113221808A/en not_active Withdrawn
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835323A (en) * | 2015-05-19 | 2015-08-12 | 银江股份有限公司 | Multi-target public transport passenger flow detection method combining with electronic fence |
| CN109522854A (en) * | 2018-11-22 | 2019-03-26 | 广州众聚智能科技有限公司 | A kind of pedestrian traffic statistical method based on deep learning and multiple target tracking |
| CN110852283A (en) * | 2019-11-14 | 2020-02-28 | 南京工程学院 | Helmet wearing detection and tracking method based on improved YOLOv3 |
| CN111860282A (en) * | 2020-07-15 | 2020-10-30 | 中国电子科技集团公司第三十八研究所 | Subway section passenger flow volume statistics and pedestrian retrograde motion detection method and system |
| CN111986237A (en) * | 2020-09-01 | 2020-11-24 | 安徽炬视科技有限公司 | Real-time multi-target tracking algorithm irrelevant to number of people |
| CN112669349A (en) * | 2020-12-25 | 2021-04-16 | 北京竞业达数码科技股份有限公司 | Passenger flow statistical method, electronic equipment and storage medium |
| CN112836639A (en) * | 2021-02-03 | 2021-05-25 | 江南大学 | Pedestrian multi-target tracking video identification method based on improved YOLOv3 model |
Non-Patent Citations (1)
| Title |
|---|
| 任珈民 等: "基于YOLOv3与卡尔曼滤波的多目标跟踪算法", 《计算机应用与软件》, no. 05 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110598558B (en) | Crowd density estimation method, device, electronic equipment and medium | |
| CN109299703B (en) | Method and device for carrying out statistics on mouse conditions and image acquisition equipment | |
| CN108154110B (en) | Intensive people flow statistical method based on deep learning people head detection | |
| CN111242108B (en) | Belt transfer point coal blockage identification method based on target detection | |
| CN111160434B (en) | Training method and device for target detection model and computer readable storage medium | |
| CN111680551A (en) | Method and device for monitoring livestock quantity, computer equipment and storage medium | |
| CN101477641A (en) | Demographic method and system based on video monitoring | |
| CN111311630A (en) | Method and system for intelligently counting quantity of goods through videos in warehousing management | |
| CN113191180B (en) | Target tracking method, device, electronic equipment and storage medium | |
| CN103581620A (en) | Image processing apparatus, image processing method and program | |
| CN115880260A (en) | Method, device and equipment for detecting base station construction and computer readable storage medium | |
| WO2021022698A1 (en) | Following detection method and apparatus, and electronic device and storage medium | |
| CN116311063A (en) | Personnel fine granularity tracking method and system based on face recognition under monitoring video | |
| CN115953719A (en) | Multi-target recognition computer image processing system | |
| CN106504242A (en) | Object detection method and system | |
| CN111751279A (en) | Optical image capturing parameter adjusting method and sensing device | |
| WO2022142416A1 (en) | Target tracking method and related device | |
| CN112329495B (en) | Bar code identification method, device and system | |
| CN113221808A (en) | Dinner plate counting statistical method and device based on image recognition | |
| KR20220146225A (en) | Pig population counting and sicked pig detection system based on deep learning video analysis technology | |
| Sutopo et al. | Appearance-based passenger counting in cluttered scenes with lateral movement compensation | |
| CN115984780A (en) | Industrial solid waste warehouse-in and warehouse-out distinguishing method and device, electronic equipment and medium | |
| CN108280402B (en) | Binocular vision-based passenger flow volume statistical method and system | |
| CN116246298A (en) | Space occupation people counting method, terminal equipment and storage medium | |
| CN105975910A (en) | Technology processing method used for carrying out video identification on moving object and system thereof |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210806 |