An Adaptive Lane Line Detection Method at Night Based on Improved Clahe and Sliding Window Search

A technology of lane line detection and sliding window search, applied in character and pattern recognition, image analysis, image enhancement and other directions, can solve the problems of poor algorithm adaptability and difficult detection of curved lane lines, achieve good real-time performance and improve algorithm efficiency. Effect

CN112270690BActive Publication Date: 2022-04-26HUAIYIN INSTITUTE OF TECHNOLOGY
13 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2022-04-26

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The present invention provides an adaptive nighttime lane line detection method based on improved CLAHE and sliding window search, comprising: setting the polygonal area in the lower half of the image as the ROI area; proposing an improved CLAHE algorithm for strong illumination image enhancement, Reduce the influence of overbright areas on lane line detection; propose a DF fusion method to separate lane lines from complex backgrounds; use the optimized sliding window search algorithm to determine lane line pixel points, and use second-order functions to fit lane line boundary equations; use The inverse perspective transformation projects the lane lines to the original image to realize lane line detection. The invention can realize lane line detection and tracking under different environments such as night light, weak light, strong light and normal light, and simultaneously solve the problem of difficult detection of curved lane lines.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical fields of image processing and road safety, in particular to an adaptive nighttime lane line detection method based on improved CLAHE and sliding window search. Background technique

[0002] There are many lane line detection methods. Nima Zarbakht et al. proposed to convert the image from RGB color space to YCbCr color space and HSV color space, and use gradient detection operator to realize lane line detection. Jamel Baili et al. proposed a feature-based lane detection method, using a horizontal difference filter to simplify the edge detection process, and using an improved Hough transform to group the detected edge points into straight lines. Qian Kidd and others used the Soberx edge detection operator to detect the edge information of lane lines, and then used the improved Hough transform to detect possible lane lines on the basis of the region of interest. The above lane line detection methods are easily aff...

Examples

Embodiment Construction

[0046] The present invention will be described in further detail below in conjunction with the accompanying drawings.

[0047] A large number of parameters are involved in this embodiment, each parameter is described as follows, as shown in Table 1.

[0048] Table 1 variable description table

[0049]

[0050] like figure 1 As shown, a kind of adaptive night lane line detection method based on improved CLAHE and sliding window search that the present invention proposes, specifically comprises the following steps:

[0051] Step 1: Acquire the video image. In order to ensure the real-time performance and efficiency of the algorithm, the down-sampling algorithm is used to scale the image, and the image size is fixed. The image length range is [600, 380] pixels, and the image width range is [460, 200] pixels. After the image img.

[0052] The resolution of the video image processed by the present invention is 1280×720. In order to improve the efficiency and applicability of ...