CN108010242B - Security alarm method, system and storage medium based on video identification - Google Patents
Security alarm method, system and storage medium based on video identification Download PDFInfo
- Publication number
- CN108010242B CN108010242B CN201711175859.6A CN201711175859A CN108010242B CN 108010242 B CN108010242 B CN 108010242B CN 201711175859 A CN201711175859 A CN 201711175859A CN 108010242 B CN108010242 B CN 108010242B
- Authority
- CN
- China
- Prior art keywords
- video
- identification
- security alarm
- alarm method
- target object
- 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
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19663—Surveillance related processing done local to the camera
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/40—Monitoring or fighting invasive species
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Closed-Circuit Television Systems (AREA)
- Alarm Systems (AREA)
- Burglar Alarm Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a security alarm method based on video identification, which comprises the following steps: an acquisition step: acquiring a video shot by a camera, wherein the video comprises a plurality of frames of video images; a first marking step: carrying out feature recognition on a target object in the video image, and marking the recognized target object to obtain a marker; early warning step: and judging whether the distribution density change of the markers of the adjacent frame video images exceeds a first preset value, and if so, carrying out early warning reminding. The invention also provides a security alarm system based on video identification and a computer readable storage medium. According to the security alarm method based on video identification, whether foreign invasive species exist is judged by judging the distribution density change of the markers in the identification area, the technical problem of judging all the foreign invasive species can be solved, and the technical implementation mode is low in cost.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a security alarm method based on video recognition, electronic equipment and a storage medium.
Background
At present, poultry farms are generally arranged in places far away from villages, towns, urban areas and the like or close to mountain forests, anti-theft safety protection is performed by means of manual patrol and fences of breeding personnel, management methods for monitoring are also provided by means of a camera device, security work is performed by means of manual regular monitoring video viewing, or security measures are implemented by means of related early warning after intrusion of strange suspicious personnel or species is monitored by means of the conventional video security monitoring technology. The traditional mode of manual patrol or manual video monitoring consumes too much labor and time, the existing security technology of video monitoring and early warning needs to judge the invasion of various species, the realization mode needs to comprehensively cover all the invaded species which may attack the raised poultry, certain technical difficulty is achieved, and the realization cost is high.
Disclosure of Invention
In order to overcome the defects of the prior art, one of the objectives of the present invention is to provide a security alarm method based on video identification, which can solve the technical problem of foreign species intrusion.
The invention also aims to provide a security alarm system based on video identification, which can solve the technical problem of foreign species intrusion.
It is a further object of the present invention to provide a computer readable storage medium that solves the technical problem of foreign species intrusion.
One of the purposes of the invention is realized by adopting the following technical scheme:
a security alarm method based on video identification comprises the following steps:
an acquisition step: acquiring a video shot by a camera, wherein the video comprises a plurality of frames of video images;
a first marking step: carrying out feature recognition on a target object in the video image, and marking the recognized target object to obtain a marker;
early warning step: and judging whether the distribution density change of the markers of the adjacent frame video images exceeds a first preset value, and if so, carrying out early warning reminding.
Further, the step of obtaining, before the step of marking, further comprises the steps of:
an extraction step: extracting at least 2 key frame images at preset time intervals according to the acquired video;
a pretreatment step: and preprocessing the extracted key frame image, wherein the preprocessing comprises binarization, noise and interference removal and image normalization.
Further, the step of obtaining, before the step of marking, further comprises the steps of:
a region segmentation step: performing region segmentation on the key frame image according to the marker shot by the camera;
an identification area construction step: and taking the segmented key frame image as an identification area of a corresponding camera.
Further, the first marking step specifically includes the following sub-steps:
a characteristic obtaining step: acquiring a feature vector of a target object in the identification area;
a judging step: judging whether the similarity between the feature vector of the target object and the feature vector of the target object sample exceeds a second preset value, if so, executing a second marking step;
a second marking step: and marking the identified target object to obtain a marker.
Further, the feature obtaining step specifically includes the following substeps:
an image segmentation step: carrying out image segmentation on a target object, and segmenting the target object into a preset number of square grid areas;
a characteristic calculation step: and calculating the area density of points in each square, wherein the area density of the points is a feature vector, and the area density of the points is the ratio of the number of points in each square to the total number of points of the image.
Further, when the number of the cameras is multiple, the acquiring step specifically includes: and acquiring videos shot by each camera, wherein the videos comprise multi-frame video images.
Further, the marker is a rectangular frame, a circular frame, an identification point or a diamond frame.
Further, the first preset value is 50%.
Further, the preset time is 1 second.
The second purpose of the invention is realized by adopting the following technical scheme:
a security alarm system based on video identification comprises a camera, a marker and a data processing server; the number of the cameras and the number of the markers are multiple;
the camera is used for acquiring a video image of a shooting area;
the marker is used for providing a reference point for the region segmentation of the video image;
the data processing server is used for executing the security alarm method based on video identification.
The third purpose of the invention is realized by adopting the following technical scheme:
a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the video recognition-based security alarm method according to any one of the objects of the present invention.
Compared with the prior art, the invention has the beneficial effects that:
according to the security alarm method based on video identification, whether foreign invasive species exist is judged by judging the distribution density change of the markers in the identification area, the technical problem of judging all the foreign invasive species can be solved, and the technical implementation mode is low in cost.
Drawings
FIG. 1 is a flow chart of a security alarm method based on video identification according to the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and the detailed description, and it should be noted that any combination of the embodiments or technical features described below can be used to form a new embodiment without conflict.
The invention is composed of: the system comprises an acquisition module, an identification module, a judgment module and an early warning module, wherein the functions of the modules are as follows
An acquisition module: a plurality of cameras are arranged in each area of a farm and used for acquiring video images of raised poultry.
Identifying a model: through machine learning and recognition training, a characteristic model library of various raised poultry such as chickens and ducks is established in advance. And acquiring photos of a large number of various raised poultry at various angles and under various light rays, and performing preprocessing, feature extraction and recognition training to establish a raised poultry recognition model. The identification model establishing process comprises the following steps:
preprocessing, preprocessing a large number of images of various poultry such as chickens, ducks and the like, binarizing the images and removing interference points. And normalizing the picture by a method of centroid alignment and linear interpolation amplification, and setting the picture as a uniform specification. And the processing and identifying performance of the server on the picture is improved through preprocessing.
And a feature extraction step, namely extracting a feature vector with a certain dimension from the preprocessed poultry raising image, so that the storage capacity and the operation speed of type matching and identification are improved. The invention correspondingly adopts the combination characteristics of the structure, the appearance and the shape of each component of the body with obvious characteristics of different raised poultry according to the images of the raised poultry. Namely, the images of the raised birds are divided into 25 square grid areas of 5 × 5, and the density distribution of the middle points of each square grid is calculated to obtain 25-dimensional feature vectors.
And a step of identification training, namely a process of extracting a standard template from various images of the raised poultry in the training set and establishing a standard feature library. Each kind of raised poultry has hundreds of standard templates, after pretreatment and feature extraction, the feature vectors of various kinds of raised poultry in the training set are stored in a file, the correct values of various kinds of raised poultry need to be indicated during training, and the recognition result is corrected through repeated training. The establishment of various poultry raising identification models is completed through the steps.
A judging module: receiving a livestock and poultry video shot by an acquisition module, extracting 2 frames of key frame images every second, preprocessing a plurality of extracted images, extracting features, comparing and identifying a model, judging raised poultry contained in the images, and marking the raised poultry in the images one by using rectangular frames; continuously analyzing density change data of the marked rectangular frames in the image, calculating the similarity of the distribution matrix of the central points of the rectangular frames, and sending the data to the early warning module when the similarity is greatly changed. The determination is made for the entire image.
The embodiment provides a security alarm system based on video identification, which comprises a camera, a marker and a data processing server; the number of the cameras and the number of the markers are multiple;
the camera is used for acquiring a video image of a shooting area;
the marker is used for providing a reference point for the region segmentation of the video image;
as shown in fig. 1, the data processing server is configured to execute a security alarm method based on video identification, which includes the following steps:
s1: acquiring videos shot by each camera, wherein the videos comprise multi-frame video images; these are the basis for data judgment and are the original data sources; the content shot by different cameras is obtained, so that different positions can be monitored;
s11: extracting at least 2 key frame images at preset time intervals according to the acquired video; the preset time is 1 second. The part is mainly a basic part for subsequent judgment, if the video image is not acquired, the content of the video image cannot be analyzed, and the subsequent content can be compared by acquiring the corresponding video and the video image directly; however, the feature recognition can be better performed by acquiring the video and extracting the key frames, because if the picture is directly acquired, some fuzzy situations of the shot picture may exist, which are not beneficial to the recognition, and such situations can be effectively avoided by acquiring the key frames in the video;
s12: and preprocessing the extracted key frame image, wherein the preprocessing comprises binarization, noise and interference removal and image normalization. And preprocessing the extracted key frame image, binarizing the image and removing interference points. And normalizing the picture by a method of centroid alignment and linear interpolation amplification, and setting the picture as a uniform specification. And the processing and identifying performance of the server on the picture is improved through preprocessing.
S13: performing region segmentation on the key frame image according to the marker shot by the camera;
s14: and taking the segmented key frame image as an identification area of a corresponding camera so as to perform security early warning on the monitoring area shot by the camera correspondingly in the later early warning.
S2: carrying out feature recognition on a target object in the video image, and marking the recognized target object to obtain a marker; step S2 specifically includes the following substeps: s21: acquiring a feature vector of a target object in the identification area; step S21 specifically includes the following substeps:
an image segmentation step: carrying out image segmentation on a target object, and segmenting the target object into a preset number of square grid areas;
a characteristic calculation step: and calculating the area density of points in each square, wherein the area density of the points is a feature vector, and the area density of the points is the ratio of the number of points in each square to the total number of points of the image. And respectively extracting the features of each preprocessed key frame image, correspondingly adopting the combined features of the structure, the appearance and the shape of each component of the body of different raised poultry with obvious features according to the images of the raised poultry, namely dividing the images of the raised poultry into 25 square grid areas of 5 by 5, and calculating the density distribution of the middle points of each square grid to obtain 25-dimensional feature vectors. Comparing the recognition models, judging the raised poultry contained in each key frame image, and marking the judged raised poultry one by using a rectangular frame.
S22: judging whether the similarity between the feature vector of the target object and the feature vector of the target object sample exceeds a preset value, if so, executing the step S23;
s23: and marking the identified target object to obtain a marker. The marker is a rectangular frame, a circular frame, an identification point or a diamond frame. These are for convenience in calculating their distribution density, and it is possible to use what form of marking. When a rectangular frame and a circular frame are adopted, the center point of the rectangular frame and the circular frame is used as a calculation standard;
s3: and judging whether the distribution density change of the markers of the adjacent frame video images exceeds a first preset value, and if so, carrying out early warning reminding. The first preset value is 50%. Continuously analyzing density change data of the marked rectangular frames in the image, calculating the central point of each rectangular frame, analyzing the similarity of the central point distribution matrix of each rectangular frame, judging that the position of the raised poultry moves suddenly in a large area when the central point distribution matrix of the rectangular frames of the extracted continuous two frames of images changes obviously in a large range, and sending the judgment result to the early warning module. For example, it is determined that the image includes more than 10 center points, and the distribution density change of the center points in the current and last two frames of images exceeds 50%, that is, 50% of the center points disappear, and we consider that it has a large-scale obvious change.
The above embodiments are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby, and any insubstantial changes and substitutions made by those skilled in the art based on the present invention are within the protection scope of the present invention.
Claims (9)
1. A security alarm method based on video identification is characterized by comprising the following steps:
an acquisition step: acquiring a video shot by a camera, wherein the video comprises a plurality of frames of video images;
a first marking step: carrying out feature recognition on a target object in the video image, and marking the recognized target object to obtain a marker; the target is raised poultry; the first marking step specifically comprises the following substeps:
a characteristic obtaining step: acquiring a feature vector of a target object in the identification area;
a judging step: judging whether the similarity between the feature vector of the target object and the feature vector of the target object sample exceeds a second preset value, if so, executing a second marking step;
a second marking step: marking the identified target object to obtain a marker;
early warning step: and judging whether the distribution density change of the markers of the adjacent frame video images exceeds a first preset value, and if so, carrying out early warning reminding.
2. The video identification-based security alarm method of claim 1, wherein the step of obtaining, before the step of marking, further comprises the steps of:
an extraction step: extracting at least 2 key frame images at preset time intervals according to the acquired video;
a pretreatment step: and preprocessing the extracted key frame image, wherein the preprocessing comprises binarization, noise and interference removal and image normalization.
3. The video identification-based security alarm method according to claim 2, wherein the preprocessing step is followed by the steps of:
a region segmentation step: performing region segmentation on the key frame image according to the marker shot by the camera;
an identification area construction step: and taking the segmented key frame image as an identification area of a corresponding camera.
4. The security alarm method based on video identification as claimed in claim 3, wherein the feature obtaining step specifically comprises the following substeps:
an image segmentation step: carrying out image segmentation on a target object, and segmenting the target object into a preset number of square grid areas;
a characteristic calculation step: and calculating the area density of points in each square, wherein the area density of the points is a feature vector, and the area density of the points is the ratio of the number of points in each square to the total number of points of the image.
5. The security alarm method based on video identification as claimed in any one of claims 1 to 4, wherein when there are a plurality of cameras, the acquiring step specifically comprises: and acquiring videos shot by each camera, wherein the videos comprise multi-frame video images.
6. The video identification-based security alarm method according to any one of claims 1-4, wherein the marker is a rectangular frame, a circular frame, an identification point or a diamond frame.
7. The video identification-based security alarm method according to any one of claims 2-4, wherein the first preset value is 50%, and the preset time is 1 second.
8. A security alarm system based on video identification is characterized by comprising a camera, a marker and a data processing server; the number of the cameras and the number of the markers are multiple;
the camera is used for acquiring a video image of a shooting area;
the marker is used for providing a reference point for the region segmentation of the video image;
the data processing server is used for executing the security alarm method based on video identification according to any one of claims 1-7.
9. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program, when executed by a processor, implements a video recognition based security alarm method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175859.6A CN108010242B (en) | 2017-11-22 | 2017-11-22 | Security alarm method, system and storage medium based on video identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711175859.6A CN108010242B (en) | 2017-11-22 | 2017-11-22 | Security alarm method, system and storage medium based on video identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108010242A CN108010242A (en) | 2018-05-08 |
| CN108010242B true CN108010242B (en) | 2020-06-09 |
Family
ID=62053244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711175859.6A Active CN108010242B (en) | 2017-11-22 | 2017-11-22 | Security alarm method, system and storage medium based on video identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108010242B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118703A (en) * | 2018-07-19 | 2019-01-01 | 苏州菲丽丝智能科技有限公司 | A kind of intelligent household security system and its working method |
| CN109858441A (en) * | 2019-01-30 | 2019-06-07 | 广州轨道交通建设监理有限公司 | A kind of monitoring abnormal state method and apparatus for construction site |
| CN110359960A (en) * | 2019-07-24 | 2019-10-22 | 精英数智科技股份有限公司 | A kind of safety alarming method and equipment for the passage of coal mine elevating conveyor |
| CN112989882A (en) * | 2019-12-16 | 2021-06-18 | 陈军 | Invasive species type identification system |
| CN111488838B (en) * | 2020-04-14 | 2023-09-01 | 上海天诚比集科技有限公司 | Method for detecting intrusion of object in video detection area |
| TWI789879B (en) * | 2021-08-23 | 2023-01-11 | 行政院農業委員會畜產試驗所 | Intelligent poultry group image recognition and analysis system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839065A (en) * | 2014-02-14 | 2014-06-04 | 南京航空航天大学 | Extraction method for dynamic crowd gathering characteristics |
| CN106203357A (en) * | 2016-07-11 | 2016-12-07 | 浙江宇视科技有限公司 | The detection method of a kind of gathering of people and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7139409B2 (en) * | 2000-09-06 | 2006-11-21 | Siemens Corporate Research, Inc. | Real-time crowd density estimation from video |
-
2017
- 2017-11-22 CN CN201711175859.6A patent/CN108010242B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839065A (en) * | 2014-02-14 | 2014-06-04 | 南京航空航天大学 | Extraction method for dynamic crowd gathering characteristics |
| CN106203357A (en) * | 2016-07-11 | 2016-12-07 | 浙江宇视科技有限公司 | The detection method of a kind of gathering of people and device |
Non-Patent Citations (1)
| Title |
|---|
| 基于视频的人群异常事件监测综述;吴新宇等;《电子测量与仪器学报》;20140630;第28卷(第6期);第575-584页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108010242A (en) | 2018-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108010242B (en) | Security alarm method, system and storage medium based on video identification | |
| US10115209B2 (en) | Image target tracking method and system thereof | |
| CN104303193B (en) | Target classification based on cluster | |
| CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
| CN108268867B (en) | License plate positioning method and device | |
| CN110738127A (en) | Helmet identification method based on unsupervised deep learning neural network algorithm | |
| CN105321350B (en) | Fake-licensed car detection method and device | |
| CN111091098A (en) | Training method and detection method of detection model and related device | |
| CN111461078B (en) | Fishing preventing and monitoring method based on computer vision technology | |
| CN112528823B (en) | Method and system for analyzing batcharybus movement behavior based on key frame detection and semantic component segmentation | |
| CN110096945B (en) | Indoor monitoring video key frame real-time extraction method based on machine learning | |
| CN111723773A (en) | Remnant detection method, device, electronic equipment and readable storage medium | |
| CN112734731A (en) | Livestock temperature detection method, device, equipment and storage medium | |
| CN111723656B (en) | Smog detection method and device based on YOLO v3 and self-optimization | |
| CN115953566B (en) | Feature analysis method, system and medium for infrared thermal imager | |
| CN115035088A (en) | Helmet wearing detection method based on yolov5 and posture estimation | |
| CN111814690A (en) | Target re-identification method and device and computer readable storage medium | |
| CN113191273A (en) | Oil field well site video target detection and identification method and system based on neural network | |
| CN111798435A (en) | Image processing method, and method and system for monitoring invasion of engineering vehicle into power transmission line | |
| CN105447463B (en) | Across the camera to automatically track system that substation is identified based on characteristics of human body | |
| CN105554456A (en) | Video processing method and apparatus | |
| CN118262282A (en) | Visual image information extraction method and system for underground fully-mechanized mining face of coal mine | |
| CN117037071A (en) | Remote supervision method, device, equipment and medium for grain insect pest of grain depot | |
| CN106778675B (en) | A kind of recognition methods of target in video image object and device | |
| CN117475353A (en) | Video-based abnormal smoke identification method and system |
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 |