Agricultural park scene lane departure early warning control method and system based on end-to-end convolutional neural network

Abstract

The invention discloses an agricultural park scene lane departure early warning control method based on an end-to-end convolutional neural network. The method comprises the following steps: collectinga real-time road image, transmitting the real-time road image to an efficient semantic segmentation convolutional neural network, and performing pixel-level marking; performing region-of-interest selection on the image marked by the neural network; extracting an edge image of the road image in the region-of-interest selection by adopting differential edge detection; obtaining a coordinate point set corresponding to edge pixels in the edge image in an image coordinate system; detecting a road edge fitting straight line from the coordinate point set corresponding to the edge pixel by adopting aHough transform straight line detection algorithm; obtaining an optimal fitting straight line through screening and fusion; and according to the optimal fitting straight line result, extracting the relative distance between the vehicle body and the left and right edges of the road in the real-time road image and the characteristics of the focus of the vehicle-mounted camera to distinguish the vehicle pose state, and carrying out corresponding vehicle body adjustment to realize centered driving of the vehicle.

Description

technical field [0001] The invention relates to the technical field of early warning of lane deviation in agricultural garden scenes, in particular to an early warning control method and system for lane deviation in agricultural garden scenes based on an end-to-end convolutional neural network. Background technique [0002] At present, the labor force in modern China is gradually concentrated in the cities, while the agricultural labor force is getting less and less, and there are large areas of barren land in the countryside; this phenomenon is particularly prominent in Xinjiang, where the land is sparsely populated. The development of large-scale and centralized management requires a large number of modern agricultural machinery and equipment. However, most of the agricultural machinery and equipment in the existing technology are almost driven by humans. This driving process is a typical simple and repetitive labor. Therefore, the need to liberate manpower from agricultur...

AI Technical Summary

Problems solved by technology

In addition, the above-mentioned patented technology also has the following problems: it requires image preprocessing to enhance edge features to improve the accuracy of edge detection, preprocessing plays a decisive role in the system, and it is necessary to manually try various image preprocessing methods and adjust various preprocessing The threshold parameter, and the robustness of the fixed preprocessing method to light changes, climate changes, road occluders, etc. is difficult to achieve good results

[0007] Second, if traditional computer vision is used to extract road edge features of agricultural dirt roads, it will not only require a lot of manual testing, but also it is difficult to get a good result

Images