Lane deviation early warning control method and system for agricultural garden scenes based on end-to-end convolutional neural network
A convolutional neural network and lane deviation technology, applied in the field of early warning and control of lane deviation in agricultural garden scenes based on end-to-end convolutional neural network, can solve a large number of manual tests, difficult to obtain results, and difficult to achieve good robustness effects and other issues
Active Publication Date: 2020-11-27
成都睿芯行科技有限公司
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
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
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0059] Such as Figure 1 to Figure 3 As shown, the present embodiment provides an end-to-end convolutional neural network-based lane deviation early warning control system for agricultural garden scenes, which consists of an image acquisition module 101, an end-to-end semantic segmentation neural network 102, and post-neural network data processing module 103, state classifier 104 and bottom vehicle control module 105. The data stream consists of input monocular RGB image 111 , semantic segmentation result 112 , road edge graph 113 , line detection result set 114 , edge detection, line detection and filtering result 115 , and control signal 116 . An imaging device capable of capturing images of the current road surface and both sides of the road surface is installed on the central axis of the vehicle.
[0060] In this embodiment, an end-to-end convolutional neural network-based lane deviation early warning control method for an agricultural garden scene includes the following...
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
Login to View More Abstract
The invention discloses an end-to-end convolutional neural network-based early warning control method for lane deviation in agricultural garden scenes, which includes the following steps: collecting real-time road images, sending them into an efficient semantic segmentation convolutional neural network for pixel-level labeling; Select the region of interest from the image marked by the neural network; use differential edge detection to extract the edge image of the road image in the region of interest selection; obtain the set of coordinate points corresponding to the edge pixels in the image coordinate system in the edge image; use Hough transform The straight line detection algorithm detects the road edge fitting straight line from the set of coordinate points corresponding to the edge pixels; the optimal fitting straight line is obtained through screening and fusion; according to the optimal fitting straight line result, the real-time road image is extracted The relative distance between the car body and the left and right edges of the road and the characteristics of the focus of the on-board camera are used to distinguish the vehicle pose state, and the corresponding body adjustments are made to achieve the 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...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): B60W30/12B60W50/14G06K9/00G06K9/32G06K9/34G06K9/62G06N3/04G06N3/08
CPCB60W30/12B60W50/14G06N3/08G06V20/588G06V10/25G06V10/267G06N3/045G06F18/2431
Inventor 周军肖剑彪龙羽徐菱
Owner 成都睿芯行科技有限公司