CN111797761B - Three-stage smoke detection system, method and readable medium - Google Patents
Three-stage smoke detection system, method and readable medium Download PDFInfo
- Publication number
- CN111797761B CN111797761B CN202010633850.0A CN202010633850A CN111797761B CN 111797761 B CN111797761 B CN 111797761B CN 202010633850 A CN202010633850 A CN 202010633850A CN 111797761 B CN111797761 B CN 111797761B
- Authority
- CN
- China
- Prior art keywords
- smoke
- smoke detection
- neural network
- video
- static
- 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.)
- Active
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
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/223—Analysis of motion using block-matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Image Analysis (AREA)
- Fire-Detection Mechanisms (AREA)
Abstract
The invention relates to the technical field of smoke detection, in particular to a three-stage smoke detection system, a method and a readable medium, which comprise the following steps that S1, a video to be detected is obtained by a camera, and the video is subjected to block processing to obtain a plurality of image blocks; s2, extracting static space features from the acquired image blocks by adopting a convolutional neural network; s3, extracting dynamic time features from static space features of a plurality of continuous image blocks in the video by adopting a cyclic neural network; s4, integrating dynamic time characteristics of a plurality of image blocks, and judging whether smoke occurs in the scene; s5, outputting a result, and performing alarm processing on the area where the smoke occurs to finish detection. The invention utilizes static space characteristics such as smoke color/texture, dynamic time characteristics such as flicker/frequency and the like and comprehensive space-time motion characteristics such as scene rising, tumbling and the like, blocks the image, improves the detection resolution capability, can accurately detect small and thin smoke in early stage, greatly improves the accuracy of smoke judgment and reduces false alarm.
Description
Technical Field
The invention relates to the technical field of smoke detection, in particular to a three-stage smoke detection system, a three-stage smoke detection method and a three-stage smoke detection readable medium.
Background
At present, smoke detection based on deep learning is roughly carried out by the following two methods: (1) The method for extracting the static features based on the convolutional neural network can avoid dependence on artificial features in the traditional method, can automatically acquire high-level features which are difficult to acquire by the traditional technology, and can achieve higher recognition rate; (2) The method for extracting dynamic characteristics based on the cyclic neural network can extract the development change of fire from smoldering to explosion, can remove static approximate smoke, and can achieve higher accuracy. There are also two methods, namely, a convolutional neural network is adopted to extract the static characteristics of each frame, and then a cyclic neural network is adopted to extract the dynamic characteristics of continuous frames, so as to achieve higher accuracy. Both the two methods and the comprehensive method of the two methods are often universal methods, and the statistical accuracy of the comprehensive sample set comprising various scenes is high; but for each scene, there are also false alarms and false alarms, which brings trouble in real-world applications.
Besides static space characteristics such as color/texture and dynamic time characteristics such as flicker/frequency, the smoke has motion characteristics such as rising, tumbling and the like in a scene in comprehensive space-time, but the characteristics cannot be represented by the motion of a pixel point or a pixel block, and the motion of the pixel point or the pixel block of the whole scene must be synthesized for representing the state. Optical flow fields have been used in some studies to analyze this movement. However, for thin smoke in early stage of fire, the optical flow field is not obvious, so that the method using the optical flow field analysis is not ideal.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention discloses a three-stage smoke detection system, a three-stage smoke detection method and a three-stage smoke detection readable medium, which are used for solving the problems that besides static space characteristics such as color/texture and dynamic time characteristics such as flicker/frequency and the like, smoke has motion characteristics in comprehensive space-time aspects such as rising, tumbling and the like in a scene, and the characteristics cannot be represented by the motion of a pixel point or a pixel block, and the motion of the pixel point or the pixel block of the whole scene must be synthesized for representing the state. Optical flow fields have been used in some studies to analyze this movement. However, for thin smoke in early stage of fire, the optical flow field is not obvious, so that the effect of the method adopting the optical flow field is not ideal.
The invention is realized by the following technical scheme:
in a first aspect, the present invention discloses a three-stage smoke detection method comprising the steps of:
s1, acquiring a video to be detected by using a camera, and performing block processing on the video to obtain a plurality of image blocks;
s2, extracting static space features from the acquired image by adopting a convolutional neural network;
s3, extracting dynamic time features from static space features of a plurality of continuous image blocks in the video by adopting a cyclic neural network; the method comprises the steps of carrying out a first treatment on the surface of the
S4, synthesizing the motions of a plurality of pixel blocks by adopting a cyclic neural network, and judging whether smoke occurs in a scene;
s5, outputting a result, and performing alarm processing on the area where the smoke occurs to finish detection.
Further, in the step S1, the video is subjected to a block processing, specifically, the t frame of the video stream is divided into n×m small blocks, and each small block is a b×b three-channel image.
Further, in the step S2, a convolutional neural network is used to extract each bxb three-channel image block into an axaxq static feature map (a>1) The a×a×q static feature map is then flattened to 1×a 2 X q dimensional static feature f cnn And further obtaining the static space characteristics of the image block.
Further, in S3, c 1×a of the consecutive frames and corresponding blocks 2 X q dimensional static feature f cnn D-dimensional local dynamic characteristic h of input cyclic neural network LDCN extraction image block st And further obtaining the dynamic time characteristics of the image.
Still further, the step S4 includes the following substeps:
s4a d-dimensional local dynamic characteristics h of each image block st Through the full connection layer FN1, outputting e-dimensional local dynamic characteristics f st ；
S4b is to make the local dynamic feature f of each image block st Merging local dynamic feature matrix M of frames according to the position of extracting image blocks in the frames LDF ；
S4c flattening the local dynamic feature matrix into a 1, n multiplied by m multiplied by e dimensional feature vector f LDF ；
S4d feature vector f of successive c frames LDF Input cycleThe ring neural network JDCN extracts global dynamic characteristics h ft ；
S4e inputs global dynamic characteristics h through a fully connected network FN2 ft And inputting a result r, and judging whether smoke and fire exist in the current frame.
In a second aspect, the present invention discloses a three-stage smoke detection system comprising a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor hardware performs the three-stage smoke detection method of the first aspect.
In a third aspect, the invention discloses a readable medium comprising execution instructions which, when executed by a processor of a three-phase smoke detection system, perform the three-phase smoke detection method of the first aspect.
The invention has the beneficial effects that
The invention not only utilizes static space characteristics such as smoke color/texture, but also utilizes dynamic time characteristics such as flicker/frequency, and the like, and simultaneously utilizes the motion characteristics of scene rising, tumbling, and the like in two aspects of comprehensive space and time, and the image is segmented, so that the resolution capability of detection can be improved, the early detection of small and thin smoke is facilitated, the accuracy of smoke judgment is greatly improved, false alarm is reduced, and the invention has strong market application prospect.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic step diagram of a three-stage smoke detection method;
fig. 2 is a block diagram of a three-stage smoke detection network in accordance with an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The embodiment discloses a three-stage smoke detection method as shown in fig. 1, comprising the following steps:
s1, acquiring a video to be detected by using a camera, and performing block processing on the video to obtain a plurality of image blocks;
s2, extracting static space features from the acquired image by adopting a convolutional neural network;
s3, extracting dynamic time features from static space features of a plurality of continuous image blocks in the video by adopting a cyclic neural network; the method comprises the steps of carrying out a first treatment on the surface of the
S4, synthesizing the motions of a plurality of pixel blocks by adopting a cyclic neural network, and judging whether smoke occurs in a scene;
s5, outputting a result, and performing alarm processing on the area where the smoke occurs to finish detection.
In S1, a video with a resolution of 1024×960 is subjected to block processing, and the t frame of the video stream is divided into 16×15 small blocks, each of which is a three-channel image of 64×64.
In S2, a convolutional neural network is adopted to extract each 64×64 three-channel image block into an 8×8×2 static feature map, and then the 8×8 static feature map is flattened into 128-dimensional static features f cnn And further obtaining static spatial characteristics of the image.
In S3, 11 128-dimensional static features f of successive frames and corresponding blocks cnn 7-dimensional local dynamic characteristic h of input cyclic neural network LDCN extraction image block st And further obtaining the dynamic time characteristics of the image.
S4 comprises the following substeps:
s4a pair7-dimensional local dynamic feature h for each image block st Through the full connection layer FN1, the 2-dimensional local dynamic characteristic f is output st ；
S4b is to make the local dynamic feature f of each image block st Merging local dynamic feature matrix M of frames according to the position of extracting image blocks in the frames LDF ；
S4c, flattening the local dynamic feature matrix to form 480-dimensional feature vector f LDF ；
S4d feature vector f of 11 consecutive frames LDF The input cyclic neural network JDCN extracts global dynamic characteristics h ft ；
S4e inputs global dynamic characteristics h through a fully connected network FN2 ft And inputting a result r, and judging whether smoke and fire exist in the current frame.
The embodiment utilizes static space characteristics such as smoke color/texture and dynamic time characteristics such as flicker/frequency and comprehensive space-time motion characteristics such as scene rising and tumbling and the like, and blocks the image, so that the resolution capability of detection can be improved, small and thin smoke in early stage can be detected, the accuracy of smoke judgment is greatly improved, and false alarm is reduced.
Example 2
The embodiment discloses a method for detecting smoke in three stages: in the first stage, a convolutional neural network is adopted to extract static space features; in the second stage, extracting dynamic time features by adopting a cyclic neural network; and thirdly, synthesizing the motions of a plurality of pixel blocks by adopting a cyclic neural network, and judging whether smoke occurs in the scene.
The network structure is shown in fig. 2, and the calculation steps are as follows:
(1) Dividing the t-th frame of the video stream into n×m small blocks, each small block being a b×b three-channel image, as shown in fig. 2 at 1;
(2) The three-channel image block of each b×b is extracted into a×a×q static feature map (a>1) The a×a×q static feature map is then flattened to 1×a 2 X q dimensional static feature f cnn As shown in fig. 2, 2;
(3) C of successive frames, corresponding blocks1×a 2 X q dimensional static feature f cnn D-dimensional local dynamic characteristic h of input cyclic neural network LDCN extraction image block st As shown in fig. 2 at 3;
(4) D-dimensional local dynamic feature h for each image block st Through the full connection layer FN1, outputting e-dimensional local dynamic characteristics f st As shown at 4 in fig. 2;
(5) Local dynamic feature f of each image block st Merging local dynamic feature matrix M of frames according to the position of extracting image blocks in the frames LDF As shown at 5 in fig. 2;
(6) Flattening the local dynamic characteristic matrix into a 1, n multiplied by m multiplied by e dimensional characteristic vector f LDF As shown at 6 in fig. 2;
(7) Feature vector f for successive c frames LDF The input cyclic neural network JDCN extracts global dynamic characteristics h ft As shown at 7 in fig. 2;
(8) Global dynamic feature h is input through fully connected network FN2 ft And inputting a result r, and judging whether smoke and fire exist in the current frame or not, as shown by 8 in fig. 2.
The method comprises the steps of 2, extracting static space features by using a convolutional neural network in the first stage, 3, extracting dynamic time features by using the convolutional neural network in the second stage, and 7, analyzing the motions of a plurality of pixel blocks in a scene by using the convolutional neural network in the third stage.
Example 3
The embodiment discloses a three-stage smoke detection system, which comprises a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor hardware executes a three-stage smoke detection method.
Example 4
The present embodiment discloses a readable medium comprising execution instructions which, when executed by a processor of a three-stage smoke detection system, perform a three-stage smoke detection method.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. A three-stage smoke detection method, the method comprising the steps of:
s1, acquiring a video to be detected by using a camera, and performing block processing on the video to obtain a plurality of image blocks;
s2, extracting static space features of each acquired image block by adopting a convolutional neural network;
s3, inputting static space features of a plurality of continuous image blocks in the video into a circulating neural network LDCN to extract d-dimensional local dynamic features h of the image blocks st ；
S4a d-dimensional local dynamic characteristics h of each image block st Through the full connection layer FN1, outputting e-dimensional local dynamic characteristics f st ；
S4b is to make the local dynamic feature f of each image block st Merging local dynamic feature matrix M of frames according to the position of extracting image blocks in the frames LDF ；
S4c is to locally and dynamically feature matrix M LDF Flattening to obtain a feature vector f LDF ；
S4d feature vector f of successive c frames LDF The input cyclic neural network JDCN extracts global dynamic characteristics h ft ；
S4e inputs global dynamic characteristics h through a fully connected network FN2 ft Judging whether smoke and fire exist in the current frame according to the output result r;
and S5, performing alarm processing on the smoke generating area to finish detection.
2. The three-phase smoke detection method according to claim 1, wherein in S1, the video is subjected to a block processing, specifically, dividing the t-th frame of the video stream into n×m small blocks, each of which is a b×b three-channel image block.
3. The three-phase smoke detection method according to claim 1, wherein in S2, a convolutional neural network is used to extract each bxb three-channel image block into a x q static feature map, where a>1, then flattening the a×a×q static feature map into 1×a 2 X q dimensional static feature f cnn And further obtaining the static space characteristics of the image block.
4. The three-phase smoke detection method according to claim 1, wherein in S3, c 1 x a of consecutive frames, corresponding blocks, are used 2 X q dimensional static feature f cnn D-dimensional local dynamic characteristic h of input cyclic neural network LDCN extraction image block st And further obtaining the dynamic time characteristics of the image block.
5. A three-phase smoke detection system comprising a processor and a memory storing execution instructions, the processor hardware performing the three-phase smoke detection method of any one of claims 1 to 4 when the processor executes the execution instructions stored in the memory.
6. A readable medium comprising execution instructions which, when executed by a processor of a three-phase smoke detection system, perform the three-phase smoke detection method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010633850.0A CN111797761B (en) | 2020-07-02 | 2020-07-02 | Three-stage smoke detection system, method and readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010633850.0A CN111797761B (en) | 2020-07-02 | 2020-07-02 | Three-stage smoke detection system, method and readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111797761A CN111797761A (en) | 2020-10-20 |
| CN111797761B true CN111797761B (en) | 2023-05-16 |
Family
ID=72811163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010633850.0A Active CN111797761B (en) | 2020-07-02 | 2020-07-02 | Three-stage smoke detection system, method and readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111797761B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117931503A (en) * | 2024-03-25 | 2024-04-26 | 华能澜沧江水电股份有限公司 | Centralized control platform fault tolerance analysis system based on similar data pre-query |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715559A (en) * | 2015-03-06 | 2015-06-17 | 温州大学 | Smoke detection and fire alarm method based on trajectory identification |
| CN108428324A (en) * | 2018-04-28 | 2018-08-21 | 温州大学激光与光电智能制造研究院 | The detection device of smog in a kind of fire scenario based on convolutional network |
| CN109190543A (en) * | 2018-08-27 | 2019-01-11 | 温州大学激光与光电智能制造研究院 | The cigarette detection method of light network based on piecemeal |
| CN109272511A (en) * | 2018-08-27 | 2019-01-25 | 温州大学激光与光电智能制造研究院 | The smoke detecting apparatus of light network based on piecemeal |
| CN110503021A (en) * | 2019-08-19 | 2019-11-26 | 温州大学 | Fire hazard smoke detecting method based on time compression track characteristic identification |
| CN111310566A (en) * | 2020-01-16 | 2020-06-19 | 国网山西省电力公司电力科学研究院 | Static and dynamic multi-feature fusion mountain fire detection method and system |
-
2020
- 2020-07-02 CN CN202010633850.0A patent/CN111797761B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715559A (en) * | 2015-03-06 | 2015-06-17 | 温州大学 | Smoke detection and fire alarm method based on trajectory identification |
| CN108428324A (en) * | 2018-04-28 | 2018-08-21 | 温州大学激光与光电智能制造研究院 | The detection device of smog in a kind of fire scenario based on convolutional network |
| CN109190543A (en) * | 2018-08-27 | 2019-01-11 | 温州大学激光与光电智能制造研究院 | The cigarette detection method of light network based on piecemeal |
| CN109272511A (en) * | 2018-08-27 | 2019-01-25 | 温州大学激光与光电智能制造研究院 | The smoke detecting apparatus of light network based on piecemeal |
| CN110503021A (en) * | 2019-08-19 | 2019-11-26 | 温州大学 | Fire hazard smoke detecting method based on time compression track characteristic identification |
| CN111310566A (en) * | 2020-01-16 | 2020-06-19 | 国网山西省电力公司电力科学研究院 | Static and dynamic multi-feature fusion mountain fire detection method and system |
Non-Patent Citations (3)
| Title |
|---|
| A Slight Smoke Perceptual Network;SHENG LUO 等;《IEEE Access》;20190430;全文 * |
| 基于级联卷积神经网络的视频动态烟雾检测;陈俊周等;《电子科技大学学报》;20161130(第06期);全文 * |
| 时间压缩轨迹特征识别的火灾烟雾检测;罗胜 等;《中国图象图形学报》;20191030;第24卷(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111797761A (en) | 2020-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111027549A (en) | Method for identifying key points of women's upper clothes | |
| CN107330390B (en) | People counting method based on image analysis and deep learning | |
| Ullah et al. | Intelligent dual stream CNN and echo state network for anomaly detection | |
| CN111126379A (en) | Target detection method and device | |
| CN111639564B (en) | Video pedestrian re-identification method based on multi-attention heterogeneous network | |
| Avgerinakis et al. | Recognition of activities of daily living for smart home environments | |
| CN113591968A (en) | Infrared weak and small target detection method based on asymmetric attention feature fusion | |
| CN109859246B (en) | Low-altitude slow unmanned aerial vehicle tracking method combining correlation filtering and visual saliency | |
| CN110020658B (en) | Salient object detection method based on multitask deep learning | |
| CN111738054A (en) | Behavior anomaly detection method based on space-time self-encoder network and space-time CNN | |
| CN115761568A (en) | Kiwi detection method based on YOLOv7 network and Deepsort network | |
| CN111401267A (en) | Video pedestrian re-identification method and system based on self-learning local feature characterization | |
| Angelo | A novel approach on object detection and tracking using adaptive background subtraction method | |
| CN111797761B (en) | Three-stage smoke detection system, method and readable medium | |
| Dahirou et al. | Motion Detection and Object Detection: Yolo (You Only Look Once) | |
| CN111008601A (en) | Fighting detection method based on video | |
| CN114120076B (en) | Cross-view video gait recognition method based on gait motion estimation | |
| CN112907621B (en) | Moving object extraction method based on difference and semantic information fusion | |
| CN112255141B (en) | Thermal imaging gas monitoring system | |
| Kalboussi et al. | A spatiotemporal model for video saliency detection | |
| Zin et al. | Background modeling using special type of Markov Chain | |
| CN111860229A (en) | Intelligent abnormal behavior identification method and device and storage medium | |
| Lin et al. | Dual-mode iterative denoiser: tackling the weak label for anomaly detection | |
| Nagmode et al. | A novel approach to detect and track moving object using Partitioning and Normalized Cross Correlation | |
| US20210357629A1 (en) | Video processing apparatus and video processing method |
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 |