Unlock instant, AI-driven research and patent intelligence for your innovation.

Thrown object identification method and device and vehicle throwing drip leakage identification method and device

An identification method and technology for throwing objects, which are applied in the field of vehicle throwing drip identification and throwing object identification, can solve the problems of high noise, false alarms in identification of throwing objects, and inability to exert the advantages of deep learning, so as to improve the recognition accuracy and reduce misjudgments. Effect

Pending Publication Date: 2021-08-13
CITY CLOUD TECH HANGZHOU CO LTD
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] (1) In the context of the traffic environment, the application of the scene is more complex, and there are many problems such as illumination changes, weather changes, camera shakes, etc., resulting in more noise in the scene, resulting in higher false positives for the identification of spilled objects, which directly affects the actual application The effect; how to reduce the false alarm of spilled drip recognition is still a problem;
[0007] (2) Most of the spilled drips are small targets and the data is difficult to collect. Without sufficient data, the advantages of deep learning cannot be brought into play. The muck thrown by the engineering vehicle is very similar to the surrounding environment, which improves the accuracy of the spilled objects. The problem of accuracy is still a big challenge;
[0008] (3) In the traditional moving target detection algorithm, the extracted spatter exists in the foreground for a short time, which is not conducive to improving the recognition ability of the extracted spatter. How to prolong the residence time of the spatter in the foreground is also the key to improving the accuracy of the spatter recognition. The essential

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Thrown object identification method and device and vehicle throwing drip leakage identification method and device
  • Thrown object identification method and device and vehicle throwing drip leakage identification method and device
  • Thrown object identification method and device and vehicle throwing drip leakage identification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] figure 1 is a flow chart of a method for identifying spilled objects according to an embodiment of the present application, such as figure 1 Said, the method includes the following steps S101 to S104:

[0054] Step S101, acquiring a sequence of video frames, and extracting a moving target from the sequence of video frames;

[0055] Step S102, processing the moving target to obtain suspected spills;

[0056] Step S103, acquiring the sample data of the spilled objects, clustering the pixels of the sample data of the spilled objects, and obtaining a clustering result;

[0057] Step S104 , acquiring the pixel value of the suspected spill, identifying the suspected spill with the pixel value in the clustering result, and obtaining the spill.

[0058] In step S101, a surveillance camera may be set to photograph the vehicles on the target road section, and the captured video may be acquired. A video frame sequence is an ordered collection of images decomposed into a video ...

Embodiment 2

[0095] The embodiment of the present application provides a method for identifying vehicle spilled drips, refer to Figure 5 , the method includes steps S501 to S503:

[0096] Step S501, process the sequence of video frames by applying the method of identifying the spilled objects to obtain the scattered objects; extract the vehicle from the video frame, wherein the video frame is obtained from the sequence of video frames;

[0097] Step S502, identifying the moving direction of the vehicle in the video frame, and setting a threshold based on the moving direction;

[0098] Step S503 , acquiring the time when the spilled object and the vehicle are close together, and calculating the distance between the two at that time, and if the distance is less than the threshold, it is determined that the vehicle is involved in the case.

[0099] In step S501, the video frame refers to an image, and in this solution, the video frame may be one frame of image or multiple consecutive frames...

Embodiment 3

[0140] Based on the same technical idea, Figure 7 An example of a sprinkler identification device provided by the embodiment of the present application is shown, including:

[0141] A moving target extraction module 701, configured to acquire a sequence of video frames, and extract a moving target from the sequence of video frames;

[0142] A processing module 702, configured to process the moving target to obtain suspected spills;

[0143] A clustering module 703, configured to acquire sample data of the spilled objects, and perform clustering on the pixels of the sample data of the spilled objects to obtain a clustering result;

[0144] The spill identification module 704 is configured to acquire the pixel value of the suspected spill, identify the suspected spill with the pixel value in the clustering result, and obtain the spill.

[0145] Among them, the moving object extraction module 701 is built with an improved mixed Gaussian model, and the spillage recognition modu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a thrown object recognition method and device and a vehicle throwing leakage recognition method and device, and the vehicle throwing leakage recognition method comprises the steps: processing a video frame sequence through employing a thrown object recognition method, and obtaining a thrown object; extracting vehicles from video frames, wherein the video frames are obtained from the video frame sequence; identifying the motion direction of a vehicle in the video frame, and setting a threshold value based on the motion direction; acquiring the moment when the thrown object is close to the position of the vehicle, calculating the distance between the thrown object and the vehicle at the moment, and if the distance is smaller than the threshold value, judging that the vehicle is a involved vehicle. According to the method, the thrown object can be obtained through multiple times of screening in the video frame sequence, and the accuracy of recognizing the thrown object is improved; and filtering the thrown objects at unreasonable positions based on the moving direction of the vehicle, so a correct matching result of the vehicle and the thrown objects is obtained.

Description

technical field [0001] The present application relates to the technical fields of image processing and machine learning, in particular to methods and devices for identifying spilled objects and vehicle spilled drips. Background technique [0002] The development of cities is inseparable from the construction of infrastructure. With the rapid development of my country's economy and society, the government's requirements for urban governance are getting higher and higher. The barbaric construction vehicle transportation method has become an urgent problem in urban management today; Spilling and dripping is the most common phenomenon during the transportation of engineering vehicles. Due to irregular loading of muck and mud, accidental spilling and dripping during transportation not only causes the problems of unclean urban roads and dust, but also makes large-scale spilling easier. Causing traffic accidents, endangering the life and property safety of drivers; throwing drips has...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/38G06K9/40G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/54G06V20/52G06V10/28G06V10/30G06V10/44G06N3/048G06N3/045G06F18/23213G06F18/241
Inventor 李圣权葛俊毛云青
Owner CITY CLOUD TECH HANGZHOU CO LTD