CN109635671A - A kind of physical distribution point counting method based on video - Google Patents
A kind of physical distribution point counting method based on video Download PDFInfo
- Publication number
- CN109635671A CN109635671A CN201811396669.1A CN201811396669A CN109635671A CN 109635671 A CN109635671 A CN 109635671A CN 201811396669 A CN201811396669 A CN 201811396669A CN 109635671 A CN109635671 A CN 109635671A
- Authority
- CN
- China
- Prior art keywords
- goods
- materials
- detection
- video
- conveyer belt
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- 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
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06T2207/30242—Counting objects in image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
- Control Of Conveyors (AREA)
Abstract
The invention belongs to video surveillance applications technical fields, and in particular to a kind of physical distribution point counting method based on video.This method is the following steps are included: establish goods and materials detection algorithm;Predict the time point and location point that next target on conveyer belt occurs;Start video intrusion detection;There are goods and materials to enter and then starts goods and materials detection algorithm;Simultaneity factor can predict next target time of occurrence and position, to count;When not having goods and materials on conveyer belt, system stops count detection, continues to start video intrusion detection.The present invention has low cost and the characteristics of primary installing can be used permanently, at the same can high efficiency and with high-precision realizations physical distribution points demand, it is very convenient to use.
Description
Technical field
The invention belongs to video surveillance applications technical fields, and in particular to a kind of physical distribution points side based on video
Method.
Background technique
Currently, for the goods and materials quantity of needs, mainly still being adopted for physical distribution especially military supplies allocation and transportation aspect
The mode checked with time-consuming and laborious craft.And the gradually development scientific and technological with society, the relatively good place of some informationizations, by
Gradually start to check using Radio Frequency Identification Technology progress goods and materials, namely is penetrated by what is purchased in advance in the goods and materials surface attaching that need to be checked
Frequency identification chip, then by the goods and materials piece by piece by induction door even handheld terminal etc., to realize substance points behaviour
Make.However, above-mentioned radio frequency identification points mode seems very convenient, when practical operation, is often ineffective.To find out its cause, being
Material on hand allocation and transportation often present the features such as single lot number amount is mostly pressed for time with allocation and transportation, since each goods and materials require on surface
RF identification chip is attached, once goods and materials quantity is excessive, not only patch is time-consuming and laborious one by one, while certainly will also result in the need for shifting to an earlier date
A large amount of RF identification chip is purchased for using, so that allocation and transportation cost is sharply promoted.At the same time, RF identification chip
There is a problem that secondary use rate is low, present value obtains the RF identification chip of secondary use, be all generally be individually worth it is up to a hundred
Chip;This is because RF identification chip easily causes wafer damage once attaching during tearing off, while after tearing off
Chip is recompiled and is recycled again, it is clear that entire treatment process is extremely complicated tediously long.Above-mentioned many reasons, and cause to penetrate
Frequency identification technology can not in physical distribution field large-scale promotion basic place.
Summary of the invention
The purpose of the present invention is overcoming above-mentioned the deficiencies in the prior art, a kind of physical distribution points side based on video is provided
Method, has low cost and the characteristics of primary installing can be used permanently, at the same can high efficiency and with high-precision realization goods and materials
Points demand is allocated and transported, it is very convenient to use.
To achieve the above object, the invention adopts the following technical scheme:
A kind of physical distribution point counting method based on video, it is characterised in that the following steps are included:
1), system relies on SVM algorithm, is learnt by obtaining goods and materials sample, to establish goods and materials detection algorithm;
2) detection zone and goods and materials outbound direction line, delimited for the video area of each video camera shooting;Meanwhile passing through
System demarcates video camera, the speed of conveyer belt transport and the positional relationship of image motion is specified, in order to which system can
The time that next target on conveyer belt occurs can be predicted according to the speed combination video camera acquired image that conveyer belt is run
Point and location point;
3), start video intrusion detection, be monitored scanning;
4), when having goods and materials to enter detection zone, and meeting the direction line of goods and materials outbound, system starts goods and materials detection and calculates
Method, system carries out target detection using goods and materials detection algorithm for the picture frame that video camera obtains and sequentially counts at this time;System
10~30 pixels are reserved before and after the location point that next target of the meeting on the conveyer belt predicted occurs, when above-mentioned mesh
When marking existing time point arrival, system starts goods and materials detection algorithm automatically and carries out target detection and count;When not having on conveyer belt
When having goods and materials, system stops count detection, returns to 3) step.
Preferably, in the step 4), when goods and materials are completely into detection zone on conveyer belt, system starts to count just now
Number.
Preferably, in the step 2), the position that video camera is set up is can shoot the goods and materials packet for understanding and transmitting on conveyer belt
Subject to the overall picture of dress.
The beneficial effects of the present invention are:
1), the present invention has been effectively combined video monitoring, SVM machine learning algorithm and big data analysis technology: first
By machine learning algorithm, it is located at the goods and materials such as the photo of different angle on conveyer belt by pre-entering a large amount of goods and materials
Sample, to support and set up a set of goods and materials detection algorithm;Later, by the speed and image motion of the transport of clear conveyer belt
Positional relationship, to realize time of occurrence point and the prediction and monitoring of location point of next target;Finally, being examined by goods and materials
While method of determining and calculating carries out goods and materials points, since system has monitoring capacity to next target time of occurrence point and location point, because
This can just open algorithm when reaching at the time point that next target occurs, and carry out the cover type detection before and after its location point,
It is final to realize high efficiency and high-precision goods and materials points purpose.
It is worth noting that, when in view of practical goods and materials points, the often synchronous allocation and transportation of multiple conveyer belts.When each is defeated
When sending band to configure and executing goods and materials detection algorithm for a long time, a large amount of algorithm calculating process certainly will be to progress algorithm operation
The performance requirement of server is high.In view of this, the present invention is by increasing target prediction on the basis of goods and materials detection algorithm
Process, so that algorithm can be automatically closed in system after current goal counting;And when next target of system prediction soon appears in
Algorithm is reopened when detection zone just now, to greatly reduce the operation burden of existing server.Due to being prediction machine
System for the accuracy for ensuring video capture image, therefore each before and after future position can reserve 10~30 pixels, and
The direct cover type of system detects 10~30 pixels when time arrives.By the above-mentioned means, the present invention bears server
Carrying requirement can be effectively reduced, and finally realize low cost and efficient goods and materials are counted demand online, achieve noticeable achievement.
Detailed description of the invention
Fig. 1 is the schematic process flow diagram of goods and materials points process of the invention;
Fig. 2 is the schematic process flow diagram of goods and materials detection algorithm and target prediction process of the invention.
Specific embodiment
For ease of understanding, here in connection with Fig. 1-2, detailed process and working method of the invention are retouched further below
It states:
The present invention provides a kind of physical distribution point counting method based on video comprising goods and materials are counted processes, goods and materials
Detection algorithm obtains process and target prediction process.Wherein:
Goods and materials points process can refer to shown in Fig. 1, and the following steps are included:
1) detection zone and goods and materials outbound direction line, delimited for the video area of each video camera shooting;
2), start video intrusion detection, be monitored scanning;
3), discovery goods and materials start to transmit, and after having goods and materials to enter detection zone, and meet the direction line of goods and materials outbound, this
When system start count detection, to enter detection zone goods and materials count;
4), after count detection to no goods and materials, stop count detection, restart video intrusion detection.
In goods and materials points process, goods and materials transmission is main to carry out goods and materials transmission using conveyer belt.And the position that video camera is set up
Set the packaging overall picture for only needing to see the entire goods and materials transmitted on conveyer belt.
And it is further, goods and materials detection algorithm obtains process and target prediction process then referring to subject to Fig. 2, including with
Lower step:
A), video camera is demarcated using system, specifies the speed of conveyer belt transport and the positional relationship of image motion;
B), before algorithm operation, system is needed to obtain goods and materials sample and is learnt using SVM algorithm, to establish
Goods and materials detection algorithm;
C), when starting detection, system carries out target detection using goods and materials detection algorithm for the picture frame obtained;Goods and materials are complete
Enter to start counting after detection zone entirely;
D), the speed of conveyer belt transport and the positional relationship of image motion are specified due to above-mentioned steps a), system
The time point and location point that the next target of speed combination image prediction that can be voluntarily run according to conveyer belt occurs.In the position
Point nearby respectively increases by 10~30 pixels, practical preferably respectively to increase by 20 pixels, when the time point that next target occurs
When arrival, system automatically turns on goods and materials detection algorithm, by the above-mentioned pixel of the detection of cover type, thus the mesh automated
Mark detection and counting operation, to evade points omission problem as far as possible, it is ensured that system is captured and the accuracy of points.
Claims (3)
1. a kind of physical distribution point counting method based on video, it is characterised in that the following steps are included:
1), system relies on SVM algorithm, is learnt by obtaining goods and materials sample, to establish goods and materials detection algorithm;
2) detection zone and goods and materials outbound direction line, delimited for the video area of each video camera shooting;Meanwhile passing through system
Video camera is demarcated, the speed of conveyer belt transport and the positional relationship of image motion are specified, in order to which system being capable of basis
Conveyer belt operation speed combination video camera acquired image can predict next target on conveyer belt appearance time point and
Location point;
3), start video intrusion detection, be monitored scanning;
4), when having goods and materials to enter detection zone, and meeting the direction line of goods and materials outbound, system starts goods and materials detection algorithm,
System carries out target detection using goods and materials detection algorithm for the picture frame that video camera obtains and sequentially counts at this time;System can be
10~30 pixels are reserved before and after the location point that next target on the conveyer belt predicted occurs, when above-mentioned target goes out
When existing time point arrives, system starts goods and materials detection algorithm automatically and carries out target detection and count;When there is no object on conveyer belt
When money, system stops count detection, returns to 3) step.
2. a kind of physical distribution point counting method based on video according to claim 1, it is characterised in that: the step 4)
In, when goods and materials are completely into detection zone on conveyer belt, system starts counting just now.
3. a kind of physical distribution point counting method based on video according to claim 1 or 2, it is characterised in that: the step
It is rapid 2) in, the position that video camera is set up is subject to and can shoot the overall picture for understanding the material packing transmitted on conveyer belt.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811396669.1A CN109635671B (en) | 2018-11-22 | 2018-11-22 | Video-based material allocation and transportation point counting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811396669.1A CN109635671B (en) | 2018-11-22 | 2018-11-22 | Video-based material allocation and transportation point counting method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109635671A true CN109635671A (en) | 2019-04-16 |
CN109635671B CN109635671B (en) | 2022-12-06 |
Family
ID=66068844
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811396669.1A Active CN109635671B (en) | 2018-11-22 | 2018-11-22 | Video-based material allocation and transportation point counting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109635671B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113158800A (en) * | 2021-03-19 | 2021-07-23 | 上海云赛智联信息科技有限公司 | Enclosure intrusion hybrid detection method and enclosure intrusion hybrid detection system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017054455A1 (en) * | 2015-09-30 | 2017-04-06 | 深圳大学 | Motion target shadow detection method and system in monitoring video |
CN107705331A (en) * | 2017-10-30 | 2018-02-16 | 中原工学院 | A kind of automobile video frequency speed-measuring method based on multiple views video camera |
-
2018
- 2018-11-22 CN CN201811396669.1A patent/CN109635671B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017054455A1 (en) * | 2015-09-30 | 2017-04-06 | 深圳大学 | Motion target shadow detection method and system in monitoring video |
CN107705331A (en) * | 2017-10-30 | 2018-02-16 | 中原工学院 | A kind of automobile video frequency speed-measuring method based on multiple views video camera |
Non-Patent Citations (1)
Title |
---|
张腾云等: "神朔铁路智能视频入侵检测***设计", 《铁路计算机应用》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113158800A (en) * | 2021-03-19 | 2021-07-23 | 上海云赛智联信息科技有限公司 | Enclosure intrusion hybrid detection method and enclosure intrusion hybrid detection system |
Also Published As
Publication number | Publication date |
---|---|
CN109635671B (en) | 2022-12-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105336074A (en) | Alarm method and device | |
CN105548198B (en) | The method of complex product quality automatic detection based on cloud platform | |
US11595937B2 (en) | Systems and methods for efficient drone detection | |
US11528452B2 (en) | Indoor positioning system using beacons and video analytics | |
CN101599175A (en) | Determine the detection method and the image processing equipment of alteration of shooting background | |
CN105426828A (en) | Face detection method, face detection device and face detection system | |
CN101853576B (en) | Embedded super-speed video detection method based on FPGA (Field Programmable Gate Array) | |
CN109635671A (en) | A kind of physical distribution point counting method based on video | |
CN108353372A (en) | The system and method for determining the position of the wireless sensor node in the network architecture used for positioning with netted feature | |
CN110245592A (en) | A method of for promoting pedestrian's weight discrimination of monitoring scene | |
CN104778723A (en) | Method for performing motion detection on infrared image with three-frame difference method | |
CN110390288B (en) | Target intelligent searching, positioning and tracking system and method based on computer vision | |
CN111950491A (en) | Personnel density monitoring method and device and computer readable storage medium | |
CN110991570A (en) | Article monitoring method based on ultrahigh frequency RFID and picture recognition | |
CN110868531B (en) | Method and device for sending trigger signal | |
US20200169928A1 (en) | Measurement reporting method and apparatus | |
CN117351472A (en) | Tobacco leaf information detection method and device and electronic equipment | |
US20220123848A1 (en) | Wireless communication quality visualization system, wireless communication quality visualization device, and measurement apparatus | |
CN109051629B (en) | Method, device and system for determining falling of materials | |
CN109941672B (en) | Material deviation rectifying method and device and electronic equipment | |
CN114937218B (en) | Electric power facility inspection system and method based on image recognition | |
US20210191429A1 (en) | Flying object control apparatus | |
CN107782259B (en) | Leaf area index measurement system based on wireless sensor network | |
CN111259677A (en) | Label identification of article, image binding method and device thereof, and storage medium | |
CN114500934B (en) | Monitoring system for realizing automatic transmission of field monitoring images |
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 |