CN115205825A - Improved YOLOV 5-based traffic sign detection and identification method for driving video sequence images - Google Patents
Improved YOLOV 5-based traffic sign detection and identification method for driving video sequence images Download PDFInfo
- Publication number
- CN115205825A CN115205825A CN202210836413.8A CN202210836413A CN115205825A CN 115205825 A CN115205825 A CN 115205825A CN 202210836413 A CN202210836413 A CN 202210836413A CN 115205825 A CN115205825 A CN 115205825A
- Authority
- CN
- China
- Prior art keywords
- traffic sign
- video sequence
- driving video
- yolov5
- image
- 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
Images
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
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a traffic sign detection and identification method based on an improved YOLOV5 driving video sequence image, which comprises the following steps: constructing a YOLOV5-Light target detection model, and building a training data set of standard traffic signs; training a YOLOV5-Light target detection model by utilizing a training data set; collecting a driving video in the driving process of a vehicle, and obtaining a driving video sequence image; extracting features by combining a Graph-Based algorithm and a CFAR algorithm, and obtaining a traffic sign candidate frame in a driving video sequence image; obtaining the distance between a distance measurement camera and a traffic sign when the current driving video sequence image is acquired; under the state of collecting the current driving video sequence image, the distance measurement camera and the included angle between the traffic sign connecting line and the horizontal plane are used for obtaining the projection image of the traffic sign candidate frame of the current driving video sequence image in front of the video camera; and inputting the projection graph into the trained Yolov5-Light target detection model to obtain the recognized traffic sign.
Description
Technical Field
The invention relates to the technical field of intelligent driving, in particular to a method for detecting and identifying traffic signs based on a driving video sequence image of an improved YOLOV 5.
Background
In the field of automatic driving technology, traffic signs, pedestrians, obstacles, and vehicle detection are one of the most central technologies. Currently, the prior art also has numerous traffic signs detected. For example, a chinese invention patent having a patent publication number of "CN112149624A" and a name of "a traffic sign image processing method and apparatus" includes: acquiring an image sequence obtained by image acquisition aiming at a road environment, and extracting a traffic identification image from each image in the image sequence; comparing the traffic identification images in any two adjacent images in the image sequence to determine the traffic identification images corresponding to the same traffic identification in any two adjacent images; associating the traffic identification images corresponding to the same traffic identification in any two adjacent images to corresponding mark points; for each marker point, the following processing is performed: and determining the traffic identification image used for representing each marking point according to the sequencing positions of the plurality of traffic identification images associated with the marking points in the image sequence.
For another example, the chinese patent invention with patent publication No. CN113255578A and name "method and apparatus for identifying traffic sign, electronic device and storage medium" includes: acquiring images and point cloud data acquired in the driving process of a vehicle; detecting the image to obtain an initial detection value sequence of each traffic identification in the image, and respectively selecting a target detection value from the initial detection value sequence of each traffic identification to obtain a target detection value sequence of the traffic identification; grouping the target detection values in the target detection value sequence, and selecting point cloud data corresponding to each group from the point cloud data; determining point cloud data corresponding to each traffic identification in each group from the point cloud data corresponding to each group, and determining the outline of each traffic identification under a reference coordinate system according to the point cloud data corresponding to each traffic identification; and marking each traffic mark in the map according to the outline of each traffic mark in the reference coordinate system.
Further, as disclosed in patent publication No. CN114387581A, "method, apparatus, storage medium, and computer device for identifying vehicle surroundings", the present invention relates to a method, apparatus, and computer device for identifying vehicle surroundings, which comprises: acquiring a peripheral environment image of a target vehicle through an image acquisition component; the image acquisition assembly is arranged around the target vehicle measuring body; extracting a traffic identification image and a background color of a traffic sign from the surrounding environment image through a pre-trained traffic identification detection neural network; the background color represents the color of a non-character object in the traffic identification image; obtaining corresponding traffic identification classification according to the background color and a preset background color and traffic identification classification relation; generating a traffic identification query instruction according to the traffic identification image and the traffic identification classification; and uploading the traffic identification query instruction to a processing server, and receiving a traffic identification query result sent by the processing server.
Although the technology can collect and identify the traffic signs, the technology has the following problems:
firstly, the acquired image is over-theoretical, the actual environment is relatively complex, and the identification accuracy is low. The situations of blurring, inclination, size change and the like exist in the acquired image; for example, traffic signs may have rotational and angular offsets in complex environments. Theoretically, the traffic sign is perpendicular to the ground, and when the traffic sign is rotated in the horizontal direction or/and the longitudinal direction along with the lapse of the use period, the area of the collected image is reduced, which increases the detection difficulty; the traditional image recognition is based on a theoretical front view, and the recognition condition is more theoretical.
Secondly, the detected environment is not only a traffic sign board, and partial occlusion or other objects with highly similar colors to the traffic sign board may exist; as such, the above techniques cannot be reliably resolved.
Thirdly, in an actual environment, the positions and heights of the traffic signs are different, so that the camera is collected in a general mode, and is not collected at a specified angle, so that the camera and the traffic signs in the collected images have a certain pitching angle, and the compression of the signs has a certain influence on the identification. However, the conventional identification methods are not corrected or improved, so that the conventional identification methods are over-theoretical and have low detection and identification efficiency.
Therefore, a method for detecting and identifying traffic signs based on improved Yolov5 driving video sequence images, which has simple logic and accurate and reliable detection, is urgently needed to be provided.
Disclosure of Invention
Aiming at the problems, the invention aims to provide a method for detecting and identifying traffic signs based on driving video sequence images of improved YOLOV5, and the technical scheme adopted by the invention is as follows:
a first part:
the technology provides a traffic sign detection and identification method based on improved YOLOV5 driving video sequence images, wherein a video camera and a ranging camera are adopted to collect driving videos; the method comprises the following steps:
constructing a YOLOV5-Light target detection model, and building a training data set of standard traffic signs;
training the YOLOV5-Light target detection model by utilizing a training data set to obtain a trained YOLOV5-Light target detection model;
collecting driving videos in the driving process of a vehicle, and obtaining driving video sequence images;
adopting a Graph-Based algorithm and a CFAR algorithm to perform combined feature extraction, and obtaining a traffic sign candidate frame in a driving video sequence image;
solving the distance between the distance measurement camera and the traffic sign when the current driving video sequence image is acquired; under the state of collecting the current driving video sequence image, the distance measurement camera and the included angle between the traffic sign connecting line and the horizontal plane are used for obtaining the projection image of the traffic sign candidate frame of the current driving video sequence image in front of the video camera;
and inputting the projection graph into the trained Yolov5-Light target detection model to obtain the recognized traffic sign.
A second part:
the technology provides a device for detecting and identifying traffic signs based on improved YOLOV5 driving video sequence images, which comprises:
the method comprises the following steps of constructing a YOLOV5-Light target detection model by using a YOLOV5-Light target detection model, and training the YOLOV5-Light target detection model by using a training data set; inputting the projection drawing into a trained Yolov5-Light target detection model to obtain a recognized traffic sign;
the video camera is arranged on a running vehicle and faces to the running direction; collecting a driving video in the driving process of a vehicle;
the distance measurement camera acquires and calculates the distance between the distance measurement camera and the traffic sign in the driving video;
the traffic sign candidate frame extraction module is connected with the video camera, adopts the combination of the Graph-Based algorithm and the CFAR algorithm for feature extraction, and obtains a traffic sign candidate frame in the driving video sequence image;
and the projection drawing conversion module is connected with the distance measurement camera and the traffic sign candidate frame extraction module, and under the state of acquiring the current driving video sequence image, the projection drawing of the traffic sign candidate frame of the current driving video sequence image in front of the video camera is obtained by the included angle between the distance measurement camera and the traffic sign connecting line and the horizontal plane.
And a third part:
the technology provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize a traffic sign detection and recognition method based on improved YOLOV5 driving video sequence image.
The fourth part:
the present technology provides a computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of a method for recognizing traffic signs based on improved YOLOV5 driving video sequence image detection.
Compared with the prior art, the invention has the following beneficial effects:
(1) The method adopts a training data set of standard traffic signs to train a YOLOV5-Light target detection model and identifies detected images; the improved YOLOV5 network is trained on a training data set of the standard traffic sign, then the trained network is used as a detection network to detect the driving video images frame by frame, and finally the automatic detection and identification of the traffic sign in the driving video are realized.
(2) The method skillfully adopts the combined feature extraction of the Graph-Based algorithm and the CFAR algorithm, and obtains a traffic sign candidate frame in a driving video sequence image; the method can realize reliable selection of the candidate frame under the condition that the background and the target have no obvious boundary, and effectively solve the problem that the traffic sign part is blocked or an object with the color highly similar to that of the traffic sign board exists.
(3) According to the invention, the distance between the ranging camera and the traffic sign and the included angle between the ranging camera and the horizontal plane in the acquisition state are obtained, so that an orthographic projection image can be conveniently obtained; the traffic sign candidate frame is orthographically projected, so that the calculation workload can be reduced, and the identification accuracy can be improved;
(4) On the basis of orthographic projection, the method combines standard traffic signs to perform scaling processing so as to improve the efficiency of detection and comparison and improve the accuracy of detection;
(5) When the projection drawing is zoomed, a YOLOV5-Light target detection model is adopted for preliminary identification, and then the preliminarily identified structure is utilized for contrast zooming; and carrying out verification and identification after scaling. Therefore, the accuracy can be further improved, and the reliable detection under the complex environment can be realized
In conclusion, the method has the advantages of simple logic, accurate and reliable detection and the like, and has high practical value and popularization value in the technical field of intelligent driving.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention, and therefore should not be considered as limiting the scope of protection, and it is obvious for those skilled in the art that other related drawings can be obtained according to these drawings without inventive efforts.
FIG. 1 is a logic flow diagram of the present invention.
Detailed Description
To further clarify the objects, technical solutions and advantages of the present application, the present invention will be further described with reference to the accompanying drawings and examples, and embodiments of the present invention include, but are not limited to, the following examples. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Examples
As shown in fig. 1, the present embodiment provides a method for detecting and recognizing traffic signs based on driving video sequence images of improved YOLOV5, wherein a video camera and a distance-measuring camera are mounted on a driving vehicle, the video camera and the distance-measuring camera face to the vehicle head, and can reliably capture images of a lane and the periphery, wherein the images have traffic signs.
In this embodiment, the method includes the steps of:
the method comprises the steps of firstly, constructing a YOLOV5-Light target detection model, and building a training data set of standard traffic signs; the training process of the standard traffic sign is very convenient due to the small data volume of the standard traffic sign.
And secondly, training the Yolov5-Light target detection model by using a training data set to obtain the trained Yolov5-Light target detection model.
And thirdly, acquiring a driving video in the driving process of the vehicle and obtaining a driving video sequence image. Under severe conditions, the traffic sign in the driving video sequence image may have a certain rotation angle, inclination angle or occlusion.
Fourthly, in order to ensure reliable extraction of the traffic sign candidate frame, the embodiment adopts the Graph-Based algorithm and the CFAR algorithm to extract the combined features, and obtains the traffic sign candidate frame in the driving video sequence image. Wherein, the Graph-Based algorithm reliably carries out borderless division. In the present embodiment, the traffic sign candidate frame extraction includes the steps of:
(1) Carrying out boundary division on the traffic sign and the background on the driving video sequence image by adopting a Graph-Based algorithm to obtain a traffic sign image;
(2) And performing morphological filtering processing on the traffic sign image to obtain a traffic sign candidate frame in the driving video sequence image.
Fifthly, solving the distance between a distance measurement camera and a traffic sign when the current driving video sequence image is acquired; under the state of collecting the current driving video sequence image, the projection image of the traffic sign candidate frame of the current driving video sequence image in front of the video camera is obtained through the included angle between the distance measurement camera and the traffic sign connecting line and the horizontal plane.
In this embodiment, the distance between the ranging camera and the traffic sign and the included angle between the ranging camera and the traffic sign connecting line and the horizontal plane are obtained, and the forward selection function is adopted for projection. In addition, the traffic sign tablet is a plane or curved surface (the denoter receives bending damage), when the measuring distance, adopts the mode of averaging, and it includes:
(1) Obtaining the distance between any characteristic point in the driving video sequence image and the distance measurement camera;
(2) Summing the distances between any characteristic point and the distance measurement camera, and obtaining the average value of the distances;
(3) And taking the average value of the distances as the distance between the ranging camera and the traffic sign.
In addition, traffic signs may have angular offsets in the longitudinal plane, which offsets may be a problem with installation, and external factors may change the angle of the traffic sign over time. Then, the embodiment corrects the traffic sign candidate frame according to the inclination angle by finding the inclination angle of the traffic sign candidate frame. For example: the straight traffic board is theoretically vertical to the ground and points to the sky, if the image standard and the horizontal plane are not at an included angle of 90 degrees, the straight traffic board has an inclined angle, and correction processing can be performed on the straight traffic board in order to guarantee the detection and identification accuracy.
Sixthly, a projection graph zooming and recognizing process comprises the following steps:
(1) Collecting the height and width of the projection drawing;
(2) Inputting the projection drawing into a trained Yolov5-Light target detection model to obtain a primary traffic sign recognition image;
(3) Zooming the projection drawing according to the height and the width of the standard traffic sign corresponding to the preliminary traffic sign identification image;
(4) And inputting the zoomed projection drawing into the trained YOLOV5-Light target detection model for verification and identification to obtain a final identified traffic sign.
(5) And inputting the projection graph into the trained Yolov5-Light target detection model to obtain the recognized traffic sign.
The above-described embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention, but all changes that can be made by applying the principles of the present invention and performing non-inventive work on the basis of the principles shall fall within the scope of the present invention.
Claims (10)
1. The traffic sign detection and identification method based on the improved YOLOV5 driving video sequence image is characterized in that a video camera and a distance measuring camera are adopted to collect driving videos; the method comprises the following steps:
constructing a YOLOV5-Light target detection model, and constructing a training data set of a standard traffic sign;
training the YOLOV5-Light target detection model by using a training data set to obtain a trained YOLOV5-Light target detection model;
collecting a driving video in the driving process of a vehicle, and obtaining a driving video sequence image;
extracting features by combining a Graph-Based algorithm and a CFAR algorithm, and obtaining a traffic sign candidate frame in a driving video sequence image;
obtaining the distance between a distance measurement camera and a traffic sign when the current driving video sequence image is acquired; under the state of collecting the current driving video sequence image, the distance measurement camera and the included angle between the traffic sign connecting line and the horizontal plane are used for obtaining the projection image of the traffic sign candidate frame of the current driving video sequence image in front of the video camera;
and inputting the projection drawing into the trained Yolov5-Light target detection model to obtain the recognized traffic sign.
2. The method for detecting and identifying traffic signs based on driving video sequence images with improved Yolov5 as claimed in claim 1, further comprising: the projected image is scaled to the size of a standard traffic sign.
3. The method for detecting and identifying traffic signs based on driving video sequence images with improved YOLOV5 as claimed in claim 1, further comprising: and obtaining the inclination angle of the traffic sign candidate frame, and correcting the traffic sign candidate frame according to the inclination angle.
4. A method for detecting and identifying traffic signs based on improved YOLOV5 driving video sequence images according to claim 1, 2 or 3, wherein the step of obtaining the distance between the distance measuring camera and the traffic signs when the current driving video sequence images are acquired comprises:
obtaining the distance between any characteristic point in the driving video sequence image and the ranging camera;
summing the distances between any characteristic point and the distance measurement camera, and obtaining the average value of the distances;
and taking the average value of the distances as the distance between the ranging camera and the traffic sign.
5. The method for recognizing traffic signs based on driving video sequence image detection with improved YOLOV5 as claimed in claim 2, wherein the scaling of the projected image to the size of standard traffic signs comprises:
collecting the height and the width of the projection drawing;
inputting the projection drawing into a trained Yolov5-Light target detection model to obtain a primary traffic sign recognition image;
zooming the projection drawing according to the height and the width of the standard traffic sign corresponding to the preliminary traffic sign identification image;
and inputting the zoomed projection drawing into the trained YOLOV5-Light target detection model for verification and identification to obtain a final identified traffic sign.
6. The method for detecting and identifying traffic signs in driving video sequence images Based on improved YOLOV5 as claimed in claim 1, wherein the method for extracting features by combining a Graph-Based algorithm and a CFAR algorithm and obtaining the candidate frames of the traffic signs in the driving video sequence images comprises the following steps:
carrying out boundary division on the traffic sign and the background on the driving video sequence image by adopting a Graph-Based algorithm to obtain a traffic sign image;
and performing morphological filtering processing on the traffic sign image to obtain a traffic sign candidate frame in the driving video sequence image.
7. An apparatus for detecting and identifying traffic signs based on driving video sequence images of improved YOLOV5 is characterized by comprising:
constructing a YOLOV5-Light target detection model, and training the YOLOV5-Light target detection model by utilizing a training data set; inputting the projection drawing into a trained Yolov5-Light target detection model to obtain a recognized traffic sign;
the video camera is arranged on a running vehicle and faces to the running direction; collecting a driving video in the driving process of a vehicle;
the distance measurement camera is used for acquiring and calculating the distance between the distance measurement camera and the traffic sign in the driving video;
the traffic sign candidate frame extraction module is connected with the video camera, adopts the combination of the Graph-Based algorithm and the CFAR algorithm for feature extraction, and obtains a traffic sign candidate frame in the driving video sequence image;
and the projection drawing conversion module is connected with the distance measurement camera and the traffic sign candidate frame extraction module, and under the state of collecting the current driving video sequence image, the distance measurement camera, the traffic sign connecting line and the horizontal plane form an included angle to obtain a projection drawing of the traffic sign candidate frame of the current driving video sequence image in front of the video camera.
8. The device for detecting and identifying traffic signs based on driving video sequence images with improved Yolov5 as claimed in claim 7, further comprising:
and the image scaling processing module scales the projection image to the size of the standard traffic sign.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for recognizing traffic signs based on improved YOLOV5 driving video sequence image detection as claimed in any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method for detecting traffic signs based on driving video sequence images with improved YOLOV5 of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210836413.8A CN115205825B (en) | 2022-07-15 | 2022-07-15 | Traffic sign detection and identification method based on improved YOLOV5 driving video sequence image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210836413.8A CN115205825B (en) | 2022-07-15 | 2022-07-15 | Traffic sign detection and identification method based on improved YOLOV5 driving video sequence image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115205825A true CN115205825A (en) | 2022-10-18 |
| CN115205825B CN115205825B (en) | 2023-06-09 |
Family
ID=83581641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210836413.8A Active CN115205825B (en) | 2022-07-15 | 2022-07-15 | Traffic sign detection and identification method based on improved YOLOV5 driving video sequence image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115205825B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456483A (en) * | 2023-12-26 | 2024-01-26 | 陕西卫仕厨房灭火设备有限公司 | Intelligent traffic driving safety warning method and device based on image processing |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663385A (en) * | 2012-03-07 | 2012-09-12 | 北京空间机电研究所 | Detection method for spot target on satellite |
| CN103020623A (en) * | 2011-09-23 | 2013-04-03 | 株式会社理光 | Traffic sign detection method and equipment |
| US9147255B1 (en) * | 2013-03-14 | 2015-09-29 | Hrl Laboratories, Llc | Rapid object detection by combining structural information from image segmentation with bio-inspired attentional mechanisms |
| DE102014006488A1 (en) * | 2014-04-30 | 2015-11-05 | Mando Corporation | SYSTEM AND METHOD FOR RECOGNIZING A SPEED LIMITATION TAG THROUGH THE USE OF A FRONT CAMERA |
| US20160379055A1 (en) * | 2015-06-25 | 2016-12-29 | Kodak Alaris Inc. | Graph-based framework for video object segmentation and extraction in feature space |
| CN106295546A (en) * | 2016-08-05 | 2017-01-04 | 淮阴工学院 | A kind of road traffic sign detection based on video and recognition methods |
| CN107729903A (en) * | 2017-10-17 | 2018-02-23 | 内蒙古科技大学 | SAR image object detection method based on area probability statistics and significance analysis |
| CN108229317A (en) * | 2017-11-28 | 2018-06-29 | 浙江工业大学 | A kind of traffic lights recognition methods calculated based on conspicuousness |
| CN109583293A (en) * | 2018-10-12 | 2019-04-05 | 复旦大学 | Aircraft Targets detection and discrimination method in satellite-borne SAR image |
| US20210042531A1 (en) * | 2018-04-27 | 2021-02-11 | Shanghai Truthvision Information Technology Co., Ltd. | Systems and methods for monitoring traffic sign violation |
| CN112464911A (en) * | 2020-12-21 | 2021-03-09 | 青岛科技大学 | Improved YOLOv 3-tiny-based traffic sign detection and identification method |
| CN112699841A (en) * | 2021-01-13 | 2021-04-23 | 华南理工大学 | Traffic sign detection and identification method based on driving video |
| CN112966810A (en) * | 2021-02-02 | 2021-06-15 | 西北大学 | Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium |
| US20210407291A1 (en) * | 2020-06-29 | 2021-12-30 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method, apparatus and electronic device for early-warning |
| CN114063055A (en) * | 2021-11-17 | 2022-02-18 | 上海鹰觉科技有限公司 | Moving target detection method and implementation system for sea surface strong clutter environment |
-
2022
- 2022-07-15 CN CN202210836413.8A patent/CN115205825B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020623A (en) * | 2011-09-23 | 2013-04-03 | 株式会社理光 | Traffic sign detection method and equipment |
| CN102663385A (en) * | 2012-03-07 | 2012-09-12 | 北京空间机电研究所 | Detection method for spot target on satellite |
| US9147255B1 (en) * | 2013-03-14 | 2015-09-29 | Hrl Laboratories, Llc | Rapid object detection by combining structural information from image segmentation with bio-inspired attentional mechanisms |
| DE102014006488A1 (en) * | 2014-04-30 | 2015-11-05 | Mando Corporation | SYSTEM AND METHOD FOR RECOGNIZING A SPEED LIMITATION TAG THROUGH THE USE OF A FRONT CAMERA |
| US20160379055A1 (en) * | 2015-06-25 | 2016-12-29 | Kodak Alaris Inc. | Graph-based framework for video object segmentation and extraction in feature space |
| CN106295546A (en) * | 2016-08-05 | 2017-01-04 | 淮阴工学院 | A kind of road traffic sign detection based on video and recognition methods |
| CN107729903A (en) * | 2017-10-17 | 2018-02-23 | 内蒙古科技大学 | SAR image object detection method based on area probability statistics and significance analysis |
| CN108229317A (en) * | 2017-11-28 | 2018-06-29 | 浙江工业大学 | A kind of traffic lights recognition methods calculated based on conspicuousness |
| US20210042531A1 (en) * | 2018-04-27 | 2021-02-11 | Shanghai Truthvision Information Technology Co., Ltd. | Systems and methods for monitoring traffic sign violation |
| CN109583293A (en) * | 2018-10-12 | 2019-04-05 | 复旦大学 | Aircraft Targets detection and discrimination method in satellite-borne SAR image |
| US20210407291A1 (en) * | 2020-06-29 | 2021-12-30 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method, apparatus and electronic device for early-warning |
| CN112464911A (en) * | 2020-12-21 | 2021-03-09 | 青岛科技大学 | Improved YOLOv 3-tiny-based traffic sign detection and identification method |
| CN112699841A (en) * | 2021-01-13 | 2021-04-23 | 华南理工大学 | Traffic sign detection and identification method based on driving video |
| CN112966810A (en) * | 2021-02-02 | 2021-06-15 | 西北大学 | Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium |
| CN114063055A (en) * | 2021-11-17 | 2022-02-18 | 上海鹰觉科技有限公司 | Moving target detection method and implementation system for sea surface strong clutter environment |
Non-Patent Citations (5)
| Title |
|---|
| XUE YUAN 等: "Traffic Sign Detection via Graph-Based Ranking and Segmentation Algorithms", 《IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS: SYSTEMS 》, vol. 45, no. 12, pages 1509, XP011589978, DOI: 10.1109/TSMC.2015.2427771 * |
| 孟雨晴: "基于Ｓ Ａ Ｒ 图像的舰船目标检测算法的研究", 《中国优秀硕士学位论文全文数据库》, no. 11, pages 032 - 2 * |
| 张上 等: "基于ＹＯＬＯｖ５的轻量化交通标志检测方法", 《电子测量技术》, vol. 45, no. 8, pages 129 - 135 * |
| 彭荣鲲;周鑫;王沛;赵永辉;琚映云;: "基于区域的GLRT车辆目标检测方法", 南京航空航天大学学报, no. 06, pages 932 - 938 * |
| 王艳丽;王志林;侯宪春;聂妍;储怡;: "基于深度学习的交通标志测距方法研究", 电脑知识与技术, no. 15, pages 1 - 3 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456483A (en) * | 2023-12-26 | 2024-01-26 | 陕西卫仕厨房灭火设备有限公司 | Intelligent traffic driving safety warning method and device based on image processing |
| CN117456483B (en) * | 2023-12-26 | 2024-03-08 | 陕西卫仕厨房灭火设备有限公司 | Intelligent traffic driving safety warning method and device based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115205825B (en) | 2023-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11854272B2 (en) | Hazard detection from a camera in a scene with moving shadows | |
| US20210334557A1 (en) | Monocular cued detection of three-dimensional strucures from depth images | |
| US9053372B2 (en) | Road marking detection and recognition | |
| US9454692B2 (en) | Method for identifying and positioning building using outline region restraint of mountain | |
| CN110458161B (en) | Mobile robot doorplate positioning method combined with deep learning | |
| US9275472B2 (en) | Real-time player detection from a single calibrated camera | |
| CN108388641B (en) | Traffic facility map generation method and system based on deep learning | |
| CN104636724B (en) | A kind of quick Pedestrians and vehicles detection method of in-vehicle camera based on goal congruence | |
| CN110658539B (en) | Vehicle positioning method, device, vehicle and computer readable storage medium | |
| CN111598952A (en) | Multi-scale cooperative target design and online detection and identification method and system | |
| CN110334625A (en) | A kind of parking stall visual identifying system and its recognition methods towards automatic parking | |
| WO2021017211A1 (en) | Vehicle positioning method and device employing visual sensing, and vehicle-mounted terminal | |
| CN107886541B (en) | Real-time monocular moving target pose measuring method based on back projection method | |
| CN115205825B (en) | Traffic sign detection and identification method based on improved YOLOV5 driving video sequence image | |
| CN112197705A (en) | Fruit positioning method based on vision and laser ranging | |
| CN109409356A (en) | A kind of multi-direction Chinese print hand writing detection method based on SWT | |
| CN108416798A (en) | A kind of vehicle distances method of estimation based on light stream | |
| CN114719873A (en) | Low-cost fine map automatic generation method and device and readable medium | |
| JP2005157731A (en) | Lane recognizing device and method | |
| Chang et al. | Real-time vision-based preceding vehicle tracking and recognition | |
| CN114140794A (en) | General image correction method based on OCR | |
| CN111898552B (en) | Method and device for distinguishing person attention target object and computer equipment | |
| CN103473787B (en) | A kind of based on moving target detecting method on the bridge of space geometry relation | |
| CN104599283A (en) | Image depth improvement method for camera height recovery based on depth difference | |
| CN114926820A (en) | Inclined parking space identification method and system based on deep learning and image frame optimization |
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 |