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

Water surface floating object multi-camera real-time detection method based on SSD network

A technology for real-time detection of floating objects on the water surface, applied in the fields of image recognition and machine learning, can solve problems such as pollution, achieve real-time performance, improve scientific performance, and avoid over-fitting problems

Pending Publication Date: 2021-10-01
DUT ARTIFICIAL INTELLIGENCE INST DALIAN +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Based on the information data of the detection of floating objects on the water surface, judging whether to deal with the floating objects on the water surface can help solve the problem of water pollution

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
  • Water surface floating object multi-camera real-time detection method based on SSD network
  • Water surface floating object multi-camera real-time detection method based on SSD network
  • Water surface floating object multi-camera real-time detection method based on SSD network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The present invention will be further described below in conjunction with specific examples.

[0035] The invention provides an SSD network-based video object detection method for floating objects on the water surface, which mainly includes data collection of floating objects on the water surface, noise reduction and enhancement of the floating object data on the water surface, data labeling of the floating objects on the water surface, SSD network model training and model detection. In order to better understand the detection method of floating objects on the water surface based on the SSD network, the working principle of the SSD network is firstly explained, as shown in figure 1 Shown:

[0036]The SSD network is a One-Stage (one-stage) target detection algorithm. Its main idea is to uniformly perform dense sampling on the feature maps of multiple layers of the picture. Different scales and aspect ratios can be used for sampling, and then CNN is used to extract featur...

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 relates to the field of machine learning and image recognition, in particular to a water surface floating object multi-camera real-time detection method based on an SSD network, and the method comprises the following steps: 1, collecting water surface floating object data through video recording, camera shooting and network collection; 2, performing water surface floating object data amplification by adopting a data noise reduction and data enhancement algorithm; 3, labeling the water surface floating object data set by adopting a Labelimg tool; 4, adopting transfer learning to train the SSD network model to obtain an optimal weight model; 5, detecting a multi-camera water surface floating object target based on the SSD network optimal weight model in real time. According to the invention, multi-camera real-time detection is carried out on the water surface floaters based on the SSD network, the interference of illumination, weather and dynamic background on real-time detection can be effectively reduced, the defect of single camera detection is made up, and the requirements of real-time performance and precision are met.

Description

technical field [0001] The invention belongs to the fields of machine learning and image recognition, and relates to a multi-camera real-time detection method for water surface floating objects based on an SSD network. Background technique [0002] With the rapid development of economy and society, the pace of urbanization in my country is accelerating, and a large number of people gather in cities to live, which has caused serious impacts on the environment and ecology around the city. Many drinking water sources, urban rivers, surrounding lakes, reservoirs, etc. A large number of pollutants appear on the water surface. The floating garbage on the water surface generally cannot be dissolved and diluted naturally, and has uneven distribution. Pollution and destruction of ecological balance even threaten the safety of shipping and drinking water. How to quickly and effectively identify floating objects on the water surface and provide early warning and real-time monitoring inf...

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): G06K9/00G06K9/40G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/24G06F18/214
Inventor 彭勇陈任飞李昱欧阳文宇吴剑岳廷秀王浅宇
Owner DUT ARTIFICIAL INTELLIGENCE INST DALIAN