Objective positioning and classifying algorithm based on deep learning

Abstract

The invention discloses an objective positioning and classifying algorithm based on deep learning. The algorithm comprises the following steps: S1, inputting a picture into a first network and outputting a series of objective positioning frames and scores; S2, inputting a picture and a series of sub windows into a second network; S3, forwardly propagating the networks to a last convolution layer to generate a characteristic pattern; S4, performing coordinate transformation on the sub windows by using a zooming coefficient in order to map the coordinates to the characteristic pattern; S5, acquiring a characteristic from the characteristic pattern by using the zoomed sub windows and pooling the characteristic to a regular size; S6, classifying the pooled data to obtain a classification result and a score of an area; and S7, inputting the objective positioning frames and the classification result of the area into a classifier to carry out classification, and outputting the type and the coordinates of the objective. The algorithm may extract characteristics, classifies the objectives, performs positioning and identification by using a unified network, and greatly increases calculating speed.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a target positioning and classification algorithm based on deep learning. Background technique [0002] The deep learning algorithm effectively organizes the processing of semantic segmentation, target detection and tracking, scene understanding and analysis, etc. in the vehicle vision under the framework of the convolutional neural network, forming an overall end-to-end processing solution, and optimizing the performance of vehicle vision. The neural network model enables it to quickly, accurately and effectively complete the task of visual processing under the vehicle embedded system. [0003] Traditional methods can only perform single-target detection. If multi-target detection is required, different features and different classifiers need to be used, which increases the difficulty of designing the entire system, and different features cannot be shared among dif...

AI Technical Summary

Problems solved by technology

[0003] Traditional methods can only perform single-target detection. If multi-target detection is required, different features and different classifiers need to be used, which increases the difficulty of designing the entire system, and different features cannot be shared among different classifiers, resulting in It reduces the repeatability of calculation and cannot improve the efficiency of detection, and the generalization ability of traditional algorithms in complex scenarios is weak and cannot achieve practical value.

Images