Intelligent water surface floating object salvage system based on neural network and image recognition

Abstract

The invention provides an intelligent water surface floating object salvage system based on a neural network and image recognition, and relates to the fields of intelligent identification and automatic control. The system comprises 1) catamaran structure modeling; 2) image acquisition and processing; 3) image recognition model modeling; 4) hull control model modeling, namely determining the deflection direction at the next moment according to the comparison of coordinates of a floating object recognized by images and three bisector coordinates of a recognition area by a linear function; and 5)building of the intelligent surface floating object salvage system, namely assembling a single-chip microcomputer, a battery, an engine, a radiator, a salvage network, a wide-angle camera, a PC and other components to form an intelligent surface floating object salvage ship, and then burning a built hull control program into the single-chip microcomputer of an intelligent surface floating objectrecognition system to achieve the functions of steering, cruising, turning round and the like of the intelligent surface floating object recognition system. By adopting the system, floating objects onthe water surface can be automatically, quickly and efficiently salvaged without human control, and the independent dredging can be realized.

Description

technical field [0001] The invention belongs to neural networks and related methods of image recognition, and specifically relates to the fields of convolutional neural networks, image processing and automatic control. Background technique [0002] Convolutional Neural Network (CNN) is a feedforward neural network. Its artificial neurons can respond to surrounding units within a part of the coverage area, and it has excellent performance for large-scale image processing. It includes convolutional layers and pooling layers. [0003] Convolutional neural network is an efficient recognition method that has been developed in recent years and has attracted widespread attention. In the 1960s, Hubel and Wiesel found that its unique network structure can effectively reduce the complexity of the feedback neural network when studying the neurons used for local sensitivity and direction selection in the cat cerebral cortex, and then proposed the convolutional neural network ( Convolu...

AI Technical Summary

Problems solved by technology

Each convolutional layer in the convolutional neural network is followed by a calculation layer for local averaging and secondary extraction. This unique feature extraction structure reduces the feature resolution.

Images