New energy vehicle abnormal parking big data detection method for smart city

A technology of a new energy vehicle and a detection method, which is applied in the field of large data detection of abnormal parking of new energy vehicles, can solve the problems of short cruising range, detection of abnormal parking conditions of new energy vehicles, hindering the smooth construction of smart cities, etc., so as to improve accuracy and reduce The effect of calculation

Active Publication Date: 2022-04-12
重庆华源智禾科技有限公司
6 Cites 5 Cited by

AI-Extracted Technical Summary

Problems solved by technology

However, due to the need for charging and short cruising range, new energy vehicles often park abnormally for a long time in the same place, which significantly hinders the smooth construction of smart cities
[0003] Traditional object detection and other...
View more

Method used

In above-mentioned steps, because self-encoding belongs to the data of low latitude between different new energy vehicle images, so select the Euclidean distance processing to be able to calculate the size data of self-encoding between different new energy vehicle images well, calculation effect Intuitive and accurate.
In the above-mentioned steps, by carrying out multi-scale reconstruction to the first image to be detected and the template image with the smallest similarity distance, their similarity is calculated at the corresponding scale, and it is possible to more accurately calculate and judge whether the image to be detected contains new energy car.
[0054] In the above-mentioned steps, by building a new energy vehicle image database, it is possible to provide an original basis for subsequent recognition to determine whether a new energy vehicle is included in the image to be detected, and to improve the accuracy of judgment. And by selecting and retaining a preset number of representative new energy vehicle image template images, the comput...
View more

Abstract

The invention provides a new energy vehicle abnormal parking big data detection method oriented to a smart city. Under the condition that a new energy automobile image database is constructed, representative new energy automobile template images are selected. And photographing the target area by using an existing camera to obtain a to-be-detected image. And performing similarity calculation on the to-be-detected image and the template image, and judging whether the to-be-detected image contains the new energy automobile or not according to a similarity calculation result. In addition, a license plate area is detected and processed by using a target detection technology and a super-resolution reconstruction technology, and on the basis, license plate information is accurately identified by using a plurality of OCR technologies in a cross mutual inspection mode. And after a preset time interval, the camera photographs the same target area, and after a series of processing, whether the original new energy vehicle is still parked in the same area is judged, so that the abnormal parking condition of the new energy vehicle is accurately and comprehensively detected.

Application Domain

Data processing applicationsCharacter and pattern recognition

Technology Topic

Image databaseImage resolution +6

Image

  • New energy vehicle abnormal parking big data detection method for smart city
  • New energy vehicle abnormal parking big data detection method for smart city
  • New energy vehicle abnormal parking big data detection method for smart city

Examples

  • Experimental program(1)
  • Effect test(1)

Example Embodiment

[0051] Example
[0052] see figure 1 , figure 1 Shown is a flowchart of a smart city-oriented new energy vehicle abnormal parking large data detection method provided by an embodiment of the present invention. The method for detecting abnormal parking data of new energy vehicles for smart cities includes the following steps:
[0053] Step S1: Build a new energy vehicle image database, and select a preset number of template images of new energy vehicle images.
[0054] In the above steps, by constructing a new energy vehicle image database, an original basis can be provided for subsequent identification and judgment of whether the image to be detected contains a new energy vehicle, and the judgment accuracy can be improved. And by selecting and retaining a preset number of representative template images of new energy vehicle images, the computing resources for running and storing the entire system can be greatly reduced, and the computing speed and accuracy can be greatly improved.
[0055] Specifically, see figure 2 , figure 2 This is a specific flow chart of the steps of constructing a new energy vehicle image database and selecting a preset number of template images of new energy vehicle images in the embodiment of the present invention. The above-mentioned building a new energy vehicle image database and selecting a preset number of new energy vehicles The steps for a template image of a car image include the following steps:
[0056] Step S1-1: Collect a preset number of new energy vehicle images to construct a new energy vehicle image database.
[0057] In the above steps, images of new energy vehicles can be obtained by uploading images of various types of new energy vehicles independently, or by taking pictures of the target area with installed cameras to obtain images of new energy vehicles, and obtaining a preset amount of new energy sources when the memory of the actual system allows. Car images to ensure the richness of the database and improve the accuracy of subsequent comparisons.
[0058] Step S1-2: Perform deep self-encoding on all new energy vehicle images.
[0059] In the above steps, the most representative information in all new energy vehicle images is extracted through deep self-encoding, and the amount of information in the new energy vehicle images is compressed, and the subsequent calculation of the self-encoded Euclidean distance between different new energy vehicle images can be performed. The amount of input information is compressed, which can effectively reduce the calculation amount of Euclidean distance calculation and improve the overall calculation speed.
[0060] Step S1-3: Calculate the self-encoded Euclidean distance between images of different new energy vehicles.
[0061] In the above steps, since the self-encoding between images of different new energy vehicles belongs to low-latitude data, the Euclidean distance processing can well calculate the size data of self-encoding between images of different new energy vehicles, and the calculation effect is intuitive and accurate.
[0062] Step S1-4: When the Euclidean distances of the multiple new energy vehicle images are lower than the preset value, only the new energy vehicle images with the smallest Euclidean distance value are retained.
[0063] Step S1-5: After calculating the Euclidean distance for many times, an image of a new energy vehicle whose Euclidean distance is not lower than a preset value is selected as a template image of the image of the new energy vehicle.
[0064] In the above steps, a batch of new energy vehicle images whose Euclidean distance is not lower than a preset value are screened out, that is, the screened new energy vehicle images have large differences and can be determined to be different types of new energy vehicle images. By adjusting the size of the preset value, the difference degree of the screened new energy vehicle images can be adjusted, thereby adjusting the difference degree of the template image of the new energy vehicle image, and providing original data of different degrees of difference for subsequent calculation and determination.
[0065] Step S2: Obtain the first image to be detected at a specific time of the target area.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Similar technology patents

Classification and recommendation of technical efficacy words

  • small amount of calculation
  • High precision

License plate detection method based on deep learning

InactiveCN106022232ARapid detectionsmall amount of calculation
Owner:CHENGDU XINEDGE TECH

Vehicle path planning method based on storage unmanned vehicle

InactiveCN107037812AOptimizing Topology Nodessmall amount of calculation
Owner:NANJING UNIV OF SCI & TECH

Phase interferometer direction finding method for ambiguity resolution by extension baselines

InactiveCN102411136AImprove direction finding accuracysmall amount of calculation
Owner:UNIV OF ELECTRONIC SCI & TECH OF CHINA

Method for forecasting short-term power in wind power station

InactiveCN102102626AHigh precision
Owner:NORTH CHINA ELECTRIC POWER UNIV (BAODING) +1

Numerical control machine tool wear monitoring method

InactiveCN102091972AHigh precisionReal-time monitoring of tool wear
Owner:HUAZHONG UNIV OF SCI & TECH +1

Advertisement recommendation method and apparatus

ActiveCN104965890AHigh precisionPrecise screening
Owner:SHENZHEN TENCENT COMP SYST CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products