Rainfall type identification method using camera video image

A video image and type recognition technology, applied in image enhancement, image analysis, image data processing, etc., to achieve good classification effect

Pending Publication Date: 2022-05-10
邹明忠
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Using image classification can overcome the problem of semantic feature extraction on the image surface, but it often needs to classify a certain data set

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
  • Rainfall type identification method using camera video image
  • Rainfall type identification method using camera video image
  • Rainfall type identification method using camera video image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention will be further described below in conjunction with the accompanying drawings.

[0033] as attached figure 1 to attach image 3 Shown, a kind of rainfall type recognition method utilizing camera video image, the present invention utilizes the image that the camera that can see everywhere on the road is taken as input, establishes rainfall category classification model by deep learning classification algorithm, comprises the following steps:

[0034] S1: Extract key frames from video images captured by several groups of cameras;

[0035] S2: Perform data processing on the key frame image in the camera video image in step S1;

[0036] S3: form a training set;

[0037] S4: Use the deep learning classification algorithm to establish a rainfall classification model, and identify different rainfall types through the rainfall classification model.

[0038] When extracting key frames from the camera video, the video captured by the camera is divided in...

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 discloses a rainfall type identification method by using camera video images, which comprises the following steps: extracting key frames from video images shot by a plurality of groups of cameras, performing data processing on key frame images in the camera video images in the step S1, forming a training set, establishing a rainfall classification model by using a deep learning classification algorithm, and identifying rainfall types by using the rainfall classification model. And identifying different rainfall types through the rainfall classification model. On the basis that images shot by cameras which can be seen everywhere on a road are processed, the defects that classification is not timely and the identification position is not specific in a traditional mode are overcome. And a plurality of low-level features can be combined to form more complex high-level feature representation by using deep learning, so that better classification of the images is realized.

Description

technical field [0001] The invention belongs to the field of ground meteorological detection, in particular to a method for identifying rainfall types using video images of a camera. Background technique [0002] Rainfall is an important part of the division of labor in the ecological cycle, and it has an important impact on agriculture, transportation, and travel activities. Under different rainfall types, the shape and scale of raindrops are different, and the impact on soil, atmosphere, and wireless communication is also different. Therefore, it is important to distinguish rainfall types. At present, the identification of rainfall type is mainly based on the change law of rainfall intensity, weather radar volume scan data, dual polarization Doppler radar polarization parameters and DSD data of raindrop spectrometer. According to the variation law of rainfall intensity and the volume scan data method of weather radar, the method is relatively simple, but it does not fully...

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
IPC IPC(8): G06V20/40G06K9/62G06N3/04G06N3/08G06T5/00G06V10/774G06V10/764G06V10/82
CPCG06N3/08G06T2207/10016G06T2207/20081G06T2207/20084G06N3/045G06F18/241G06F18/214G06T5/80G06T5/70
Inventor 邹明忠钱彬源朱珉吉
Owner 邹明忠
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap