Structured light strip center extraction method for asphalt pavement image

Abstract

The invention discloses a method for extracting the center of a structured light strip of an asphalt pavement image. The method comprises steps of dividing an image through employing a light strip gray threshold, cutting an ROI (region of interest) of the light strip of the image, and carrying out the subsequent extraction algorithm in the ROI; adopting an unbiased extraction method of a Steger curve stripe center to obtain a light bar sub-pixel center point; a local outlier factor (LOF) detection method being used for the light bar center point extracted by the Steger, outliers being found and removed by calculating the local density of each point, and the accurate light bar center point being obtained. According to the structured light strip center extraction method for the asphalt pavement image, the light strip center extraction effect is obviously improved, interference is removed for the characteristics of noise points, unnecessary convolution calculation is avoided, calculationefficiency is high, and the method has high universality and robustness and is suitable for large-scale popularization and application. The method is applied to pavement surface three-dimensional scanning and apparent reconstruction, and extraction precision and recognition efficiency of the apparent information of the asphalt pavement are improved.

Description

technical field [0001] The invention relates to a method for extracting the center of a structured light strip of an asphalt pavement image, and belongs to the field of road engineering and the technical field of image processing. Background technique [0002] As of the end of 2019, the country's highway mileage has reached 5.0125 million kilometers, of which the total mileage of expressways has reached 149,600 kilometers. As the country's most basic, most concerned and heavily invested infrastructure project, the quality of roads affects the economic development and people's daily life in various places. Therefore, the importance and urgency of road inspection and maintenance management have become increasingly prominent. At present, with the development of industrial cameras, image processing and other technologies and the improvement of computing power, the road surface 3D scanning and acquisition equipment has become mature, but the detection of the depth of the structur...

AI Technical Summary

Problems solved by technology

[0004] (1) The traditional light strip center extraction methods mainly include: extreme value method, threshold method, center of gravity method, curve fitting method, edge method, geometric center method, etc. For grayscale curve fitting, the extracted center point is a pixel-level coordinate, which has large errors and low precision. Due to the influence of factors such as the environment, the complexity of the device itself and the measured object, it is difficult for these light bar center extraction methods to meet the requirements simultaneously. More universal, real-time and precise measurement requirements

[0005] (2) The most representative sub-pixel-level center extraction method is the "unbiased extraction method for the center of curved stripes" proposed by Dr. Steger. For the characteristics of stripes (such as blood vessels, roads, etc.) in medical images and satellite images, according to Taylor polynomial expansion, to obtain the sub-pixel center of the gray distribution, has strong universality and robustness, and the extraction accuracy is high, but because it uses the Gaussian kernel image convolution of the large template, the calculation efficiency is low , and when applied to images with complex gray levels of asphalt pavement, there will be noise

[0006] Among the existing methods, the traditional method has higher requirements on the road image quality and is greatly disturbed by the external environment. However, the asphalt pavement detection in the actual project is not obtained under the same lighting conditions (day / night), (sunny day / Cloudy), the Steger method has obvious advantages in accuracy, but the calculation efficiency of Gaussian convolution in a large range is low, and noise will be generated in places with large local gray levels at the edge of the light strip, which will affect the efficiency and accuracy of center extraction

Images