CN105373782A - Method of automatically recognizing hazardous chemical vehicle from image or video - Google Patents
Method of automatically recognizing hazardous chemical vehicle from image or video Download PDFInfo
- Publication number
- CN105373782A CN105373782A CN201510782360.6A CN201510782360A CN105373782A CN 105373782 A CN105373782 A CN 105373782A CN 201510782360 A CN201510782360 A CN 201510782360A CN 105373782 A CN105373782 A CN 105373782A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- harmful influence
- image
- object detector
- hazardous chemical
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method of automatically recognizing a hazardous chemical vehicle from an image or video. The method comprises the following steps: (1) a to-be-detected vehicle image is acquired through a monitoring device; (2) whether a hazardous chemical signboard exists in the vehicle image is detected, if yes, the vehicle is directly determined to be a hazardous chemical vehicle, or otherwise, the next step is carried out; (3) through a vehicle type comparison mode, whether the vehicle without the hazardous chemical signboard is a tank wagon, if yes, the vehicle is determined to be a suspected hazardous chemical vehicle, or otherwise, the next step is carried out; and (4) through an object detector, whether a tank object exists in a carriage range of the vehicle is detected, if the tank object exists, the vehicle is determined to be a suspected hazardous chemical vehicle, or otherwise, the vehicle is not a hazardous chemical vehicle. Automatic and quick hazardous chemical vehicle recognition can be effectively realized, the management means on the hazardous chemical vehicle is enhanced, and an important role is played on hazardous chemical vehicle safety accident prevention.
Description
Technical field
The present invention relates to a kind of harmful influence vehicle identification method, specifically, relate to a kind of method automatically identifying harmful influence vehicle from picture or video.
Background technology
Along with China's rapid economic development, road and logistics transportation increase sharply, traffic safety and the unimpeded work becoming the most important thing; especially harmful influence; because it is inflammable, explosive, easily poisoning, easily pollute, in transportation once have an accident, consequence is hardly imaginable.Harmful influence vehicle transport in recent years; due to long distance, trans-regional, monitoring improvement is got up very difficult, and harmful influence is once set out on a journey; traffic police has just had road surface transform responsibility; and on-site law-enforcing, due to reasons such as region are wide, route is many, police strength is not enough, must fall flat; add some illegal vehicle; extremely draw and run chaotically, ignore the possible consequences, bring great risk to people's life's property.
Present harmful influence vehicle management, mainly uses the logistics vehicles track and localization based on GPS, the goods information based on RFID to follow the tracks of and to review etc. logistics regulatory format.This improves the security of harmful influence logistics system to a certain extent.But be that gps system or rfid system all can only manage the harmful influence vehicle in establishment, still have no way of supervising to the outer harmful influence vehicle not installing associated assay devices of establishment, detection leakage phenomenon is very serious.
Summary of the invention
The object of the present invention is to provide a kind of method automatically identifying harmful influence vehicle from picture or video, install checkout equipment without the need to vehicle, third party directly can realize the identification of harmful influence vehicle.
To achieve these goals, the technical solution used in the present invention is as follows:
From picture or video, automatically identify a method for harmful influence vehicle, comprise the following steps:
(1) vehicle image to be detected is obtained by watch-dog;
(2) detect in vehicle image whether there is harmful influence sign board, have, then directly determine that this vehicle is harmful influence vehicle, otherwise, then perform next step;
(3) judge whether the vehicle without harmful influence sign board is canned lorry by vehicle way of contrast, if so, then determine that this vehicle is doubtful harmful influence vehicle, otherwise, then perform next step;
(4) by whether there is canned object within the scope of the compartment of object detector detection vehicle, if there is canned object, then doubtful harmful influence vehicle is defined as, otherwise, be then non-harmful influence vehicle.
Further, judge whether in described step (2) that the method that there is harmful influence sign board is specific as follows:
(2a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in vehicle image;
(2b) edge extracting is carried out to the image-region in vehicle range;
(2c) the edge line segment extracted is analyzed, respectively for headstock picture and tailstock image detect triangle or argyle design;
(2d) utilize SVM classifier to classify to the triangle detected in vehicle range or argyle design, be confirmed whether as harmful influence sign board.
Again further, described SVM classifier is obtained by harmful influence sign board picture sample training.
Further, judge whether in described step (3) that the method that there is canned lorry is specific as follows:
(3a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in image;
(3b) filtration treatment and normalized are carried out to vehicle image, to obtain the vehicle image after normalized;
(3c) proper vector of the vehicle image after normalized is obtained;
(3d) the corresponding proper vector of the proper vector of the vehicle image after normalized to the vehicle sample in database is compared, with obtain with normalized after the immediate database of vehicle image in K vehicle sample;
(3e) determine whether this vehicle to be detected is canned lorry according to the type demarcation of the K in database vehicle sample.
Preferably, the object detector in described step (4) is based on the object detector of HOG feature or the object detector based on Haar feature.
Compared with prior art, the present invention has following beneficial effect:
The present invention is based on existing electronic traffic watch-dog; with the sign board on vehicle, vehicle interior article and vehicle for detected object; judged by contrast; have effectively achieved harmful influence vehicle automatic, identify fast; enhance the ladder of management to harmful influence vehicle; improve accuracy rate and the detection efficiency of harmful influence vehicle detection, to the strick precaution of harmful influence vehicle safety accident, there is very great effect.
Accompanying drawing explanation
Fig. 1 is schematic flow sheet of the present invention.
Fig. 2 is a kind of vehicle schematic diagram being loaded with risk identification board in the present invention-embodiment.
Fig. 3 is the schematic diagram of a kind of canned lorry in the present invention-embodiment.
Fig. 4 is the schematic diagram of the canned object of a kind of canned charge of trucks in the present invention-embodiment.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described, and embodiments of the present invention include but not limited to the following example.
Embodiment
As shown in Figure 1, the method automatically identifying harmful influence vehicle from picture or video disclosed by the invention, comprises the following steps:
(1) vehicle image to be detected is obtained by watch-dog;
(2) detect in vehicle image whether there is harmful influence sign board, have, then directly determine that this vehicle is harmful influence vehicle, otherwise, then perform next step;
(3) judge whether the vehicle without harmful influence sign board is canned lorry by vehicle way of contrast, if so, then determine that this vehicle is doubtful harmful influence vehicle, otherwise, then perform next step;
(4) by whether there is canned object within the scope of the compartment of object detector detection vehicle, if there is canned object, then doubtful harmful influence vehicle is defined as, otherwise, be then non-harmful influence vehicle.
Specifically, judge whether in described step (2) that the method that there is harmful influence sign board is specific as follows:
(2a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in vehicle image;
(2b) edge extracting is carried out to the image-region in vehicle range;
(2c) the edge line segment extracted is analyzed, respectively for headstock picture and tailstock image detect triangle or argyle design;
(2d) utilize SVM classifier to classify to the triangle detected in vehicle range or argyle design, be confirmed whether as harmful influence sign board.
Wherein, described SVM classifier is obtained by harmful influence sign board picture sample training.
As shown in Figure 2, the common vehicle being loaded with harmful influence is provided with hazard identification board, and its shape is generally triangle or rhombus.Triangle hazard identification board is arranged on headstock, and rhombus dangerous mark board is arranged on the tailstock, therefore, when detecting, utilizes different detection algorithm to detect respectively for headstock and tailstock picture.Because the algorithm detecting hazard identification board is ripe, therefore this algorithm is not described in detail at this.
Judge whether in described step (3) that the method that there is canned lorry is specific as follows:
(3a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in image;
(3b) filtration treatment and normalized are carried out to vehicle image, to obtain the vehicle image after normalized;
(3c) proper vector of the vehicle image after normalized is obtained;
(3d) the corresponding proper vector of the proper vector of the vehicle image after normalized to the vehicle sample in database is compared, with obtain with normalized after the immediate database of vehicle image in K vehicle sample;
(3e) determine whether this vehicle to be detected is canned lorry according to the type demarcation of the K in database vehicle sample.
The vehicle sample of canned lorry rear view as shown in Figure 3, it should be noted that, vehicle sample shown in Fig. 3 is only a kind of example, in reality, also there is the canned lorry of other patterns, in database of the present invention, will the various canned lorry pattern in existing vehicle be collected, and carry out real-time update, to ensure the comprehensive of vehicle sample, guarantee that the canned goods car test is surveyed accurate, at utmost avoid undetected problem.
Preferably, the object detector in described step (4) is based on the object detector of HOG feature or the object detector based on Haar feature.
Fig. 4 is the schematic diagram being loaded with canned object in a kind of canned lorry.
In the present invention; finally be detected as doubtful harmful influence vehicle owing to belonging to canned vehicle or canned object; automatically cannot be detected by instrument; because the automatic testing result that the present invention realizes can only determine doubtful harmful influence vehicle; as needs finally determine whether as harmful influence vehicle; then need to carry out hand inspection by law enforcement agency, to this, do not describe in detail herein.
The present invention is by detecting hazard identification board, canning cart and canned object, contrast respectively; analyze and judge; automatically harmful influence vehicle is identified; not only detect wide; and efficiency is fast, accuracy rate is high, effectively can strengthen the ladder of management of harmful influence vehicle, effectively strick precaution illegal vehicle extremely draws and runs chaotically; reduce the probability that harmful influence vehicle has an accident, there is very high practical value.
Above-described embodiment is only the preferred embodiments of the present invention, not limiting the scope of the invention, as long as adopt design concept of the present invention, and the change carried out non-creativeness work on this basis and make, all should belong within protection scope of the present invention.
Claims (5)
1. from picture or video, automatically identify a method for harmful influence vehicle, it is characterized in that, comprise the following steps:
(1) vehicle image to be detected is obtained by watch-dog;
(2) detect in vehicle image whether there is harmful influence sign board, have, then directly determine that this vehicle is harmful influence vehicle, otherwise, then perform next step;
(3) judge whether the vehicle without harmful influence sign board is canned lorry by vehicle way of contrast, if so, then determine that this vehicle is doubtful harmful influence vehicle, otherwise, then perform next step;
(4) by whether there is canned object within the scope of the compartment of object detector detection vehicle, if there is canned object, then doubtful harmful influence vehicle is defined as, otherwise, be then non-harmful influence vehicle.
2. a kind of method automatically identifying harmful influence vehicle from picture or video according to claim 1, is characterized in that, judge whether that the method that there is harmful influence sign board is specific as follows in described step (2):
(2a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in vehicle image;
(2b) edge extracting is carried out to the image-region in vehicle range;
(2c) the edge line segment extracted is analyzed, respectively for headstock picture and tailstock image detect triangle or argyle design;
(2d) utilize SVM classifier to classify to the triangle detected in vehicle range or argyle design, be confirmed whether as harmful influence sign board.
3. a kind of method automatically identifying harmful influence vehicle from picture or video according to claim 2, it is characterized in that, in described step (2d), SVM classifier is obtained by harmful influence sign board picture sample training.
4. a kind of method automatically identifying harmful influence vehicle from picture or video according to claim 1, is characterized in that, judge whether that the method that there is canned lorry is specific as follows in described step (3):
(3a) object detector based on HOG feature or the object detector based on Haar feature is used to position the vehicle in image;
(3b) filtration treatment and normalized are carried out to vehicle image, to obtain the vehicle image after normalized;
(3c) proper vector of the vehicle image after normalized is obtained;
(3d) the corresponding proper vector of the proper vector of the vehicle image after normalized to the vehicle sample in database is compared, with obtain with normalized after the immediate database of vehicle image in K vehicle sample;
(3e) determine whether this vehicle to be detected is canned lorry according to the type demarcation of the K in database vehicle sample.
5. a kind of method automatically identifying harmful influence vehicle from picture or video according to claim 1, it is characterized in that, the object detector in described step (4) is based on the object detector of HOG feature or the object detector based on Haar feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782360.6A CN105373782A (en) | 2015-11-16 | 2015-11-16 | Method of automatically recognizing hazardous chemical vehicle from image or video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782360.6A CN105373782A (en) | 2015-11-16 | 2015-11-16 | Method of automatically recognizing hazardous chemical vehicle from image or video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105373782A true CN105373782A (en) | 2016-03-02 |
Family
ID=55375966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510782360.6A Pending CN105373782A (en) | 2015-11-16 | 2015-11-16 | Method of automatically recognizing hazardous chemical vehicle from image or video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105373782A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107016364A (en) * | 2017-04-01 | 2017-08-04 | 南京邮电大学 | A kind of express delivery location acquiring method based on image recognition |
| CN107146062A (en) * | 2017-05-31 | 2017-09-08 | 合肥亿迈杰软件有限公司 | A kind of logistics vehicles turnover community's control system based on big data |
| CN107240176A (en) * | 2017-05-31 | 2017-10-10 | 合肥亿迈杰软件有限公司 | A kind of district vehicles turnover management method of feature based identification |
| CN111814562A (en) * | 2020-06-11 | 2020-10-23 | 浙江大华技术股份有限公司 | Vehicle identification method, vehicle identification model training method and related device |
| CN111898502A (en) * | 2020-07-20 | 2020-11-06 | 北京格灵深瞳信息技术有限公司 | Dangerous goods vehicle identification method and device, computer storage medium and electronic equipment |
| CN111931545A (en) * | 2019-05-13 | 2020-11-13 | 杭州海康威视数字技术股份有限公司 | Number plate identification method and device |
| CN112068520A (en) * | 2020-09-14 | 2020-12-11 | 苏州颂康智能科技有限公司 | Remote bridge hazardous chemical substance accident real-time prevention and control system based on 5G |
| CN113111884A (en) * | 2021-03-26 | 2021-07-13 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method for special vehicle for hazardous chemical transportation |
| CN113469158A (en) * | 2021-09-06 | 2021-10-01 | 智广海联(天津)大数据技术有限公司 | Method and system for identifying illegal hazardous chemical substance transport vehicle based on convolutional neural network |
| CN113723258A (en) * | 2021-08-24 | 2021-11-30 | 广州邦讯信息系统有限公司 | Dangerous goods vehicle image identification method and related equipment thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015041389A1 (en) * | 2013-09-23 | 2015-03-26 | 배문준 | Vehicle control system for dangerous article management |
| CN104715231A (en) * | 2013-12-11 | 2015-06-17 | 深圳市朗驰欣创科技有限公司 | Method and device for monitoring dangerous goods transportation vehicles at traffic intersection |
| CN104732245A (en) * | 2015-04-14 | 2015-06-24 | 万里运业股份有限公司 | Identification and pre-warning method of hazardous article conveying vehicle appearing in front of coach bus |
-
2015
- 2015-11-16 CN CN201510782360.6A patent/CN105373782A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015041389A1 (en) * | 2013-09-23 | 2015-03-26 | 배문준 | Vehicle control system for dangerous article management |
| CN104715231A (en) * | 2013-12-11 | 2015-06-17 | 深圳市朗驰欣创科技有限公司 | Method and device for monitoring dangerous goods transportation vehicles at traffic intersection |
| CN104732245A (en) * | 2015-04-14 | 2015-06-24 | 万里运业股份有限公司 | Identification and pre-warning method of hazardous article conveying vehicle appearing in front of coach bus |
Non-Patent Citations (1)
| Title |
|---|
| 李星等: "基于HOG特征和SVM的前向车辆识别方法", 《计算机科学》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107016364A (en) * | 2017-04-01 | 2017-08-04 | 南京邮电大学 | A kind of express delivery location acquiring method based on image recognition |
| CN107146062A (en) * | 2017-05-31 | 2017-09-08 | 合肥亿迈杰软件有限公司 | A kind of logistics vehicles turnover community's control system based on big data |
| CN107240176A (en) * | 2017-05-31 | 2017-10-10 | 合肥亿迈杰软件有限公司 | A kind of district vehicles turnover management method of feature based identification |
| CN111931545B (en) * | 2019-05-13 | 2023-10-31 | 杭州海康威视数字技术股份有限公司 | License plate identification method and device |
| CN111931545A (en) * | 2019-05-13 | 2020-11-13 | 杭州海康威视数字技术股份有限公司 | Number plate identification method and device |
| CN111814562A (en) * | 2020-06-11 | 2020-10-23 | 浙江大华技术股份有限公司 | Vehicle identification method, vehicle identification model training method and related device |
| CN111898502A (en) * | 2020-07-20 | 2020-11-06 | 北京格灵深瞳信息技术有限公司 | Dangerous goods vehicle identification method and device, computer storage medium and electronic equipment |
| CN112068520A (en) * | 2020-09-14 | 2020-12-11 | 苏州颂康智能科技有限公司 | Remote bridge hazardous chemical substance accident real-time prevention and control system based on 5G |
| CN113111884A (en) * | 2021-03-26 | 2021-07-13 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method for special vehicle for hazardous chemical transportation |
| CN113111884B (en) * | 2021-03-26 | 2024-05-24 | 沈阳天眼智云智能技术研究院有限公司 | Video detection method of special dangerous chemical transportation vehicle |
| CN113723258A (en) * | 2021-08-24 | 2021-11-30 | 广州邦讯信息系统有限公司 | Dangerous goods vehicle image identification method and related equipment thereof |
| CN113723258B (en) * | 2021-08-24 | 2024-05-28 | 广州邦讯信息系统有限公司 | Dangerous goods vehicle image recognition method and related equipment thereof |
| CN113469158A (en) * | 2021-09-06 | 2021-10-01 | 智广海联(天津)大数据技术有限公司 | Method and system for identifying illegal hazardous chemical substance transport vehicle based on convolutional neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105373782A (en) | Method of automatically recognizing hazardous chemical vehicle from image or video | |
| CN103512762B (en) | Image processing method, device and train failure detection system | |
| CN107273802B (en) | Method and device for detecting fault of brake shoe drill rod ring of railway train | |
| CN106448181B (en) | Non-motor vehicle management method, device and system | |
| CN104751145B (en) | The SAR image electric power line detecting method that local Hough transformation optimizes with morphology | |
| CN104827963B (en) | A kind of method, control system and control device for collision prevention of vehicle intelligent early-warning | |
| US11884310B2 (en) | Systems and methods for detecting tanks in railway environments | |
| CN103390145B (en) | A kind of target area vehicle checking method and system | |
| CN105259304A (en) | On-line monitoring system and method for pollutants in vehicle tail gas | |
| CN204945122U (en) | A kind of Vehicular exhaust pollutant monitoring system | |
| CN103021182B (en) | Method and device for monitoring motor vehicle in case of regulation violation for running red light | |
| CN109166321B (en) | Road traffic vehicle monitoring method and road traffic vehicle monitoring system | |
| CN107067730A (en) | Net based on the tollgate devices about inconsistent monitoring method of car people car | |
| CN105608903A (en) | Traffic violation detection method and system | |
| RU2013105778A (en) | METHOD FOR PREPARING AND MAINTENANCE OF VEHICLES | |
| CN113903180B (en) | Method and system for detecting vehicle overspeed on expressway | |
| US20110054730A1 (en) | System and process to record and transmit inspection information | |
| Shin et al. | Enhancing Railway Maintenance Safety Using Open‐Source Computer Vision | |
| CN204341099U (en) | A kind of railway large-scale maintenance machinery runs anti-collision prewarning apparatus | |
| CN106205137A (en) | A kind of yellow mark tail gas monitors identification system automatically | |
| CN107392093B (en) | A kind of rail identifying system combined based on machine learning and gray projection algorithm | |
| CN113723258B (en) | Dangerous goods vehicle image recognition method and related equipment thereof | |
| CN103076641B (en) | A kind of safety detecting system and detection method | |
| CN203102417U (en) | Dangerous vehicle passing system | |
| TWM635693U (en) | Automatic detection device for oversized and dangerous goods vehicles |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160302 |
|
| WD01 | Invention patent application deemed withdrawn after publication |