Estimating degree of deterioration of road surface

摘要

Image analysis methods for quantifying cracks in a road surface from a road surface image recorded as a digital image, and quantify the degree of deterioration of the road surface. An object image is prepared, in which its region is divided into a plurality of pixels and the grayscale value of each of the pixels is inverted. An image analysis technique may include, on the basis of an image obtained by applying a Gabor filter to a multiple resolution image, containing images with multiple scales obtained by scaling, cracks in a road surface can be detected and distinguished from white lines and characters on the road surface. It is possible to solve various problems, in which visually detected cracks do not appear as edges and edges of white lines and characters on the road surface are detected as line segment vectors.

主权项

1. A method for detecting cracks from an object image, the method comprising the steps of:
dividing a region of the object image into a plurality of pixels; inverting a grayscale value of each of the plurality of pixels; preparing a group of filters with different rotational orientation components and applying one of the filters to the plurality of pixels; setting values of pixels with filter responses greater than or equal to a first threshold value to 1 and setting values of the other pixels to 0; applying a filter, to the plurality of pixels, wherein the filter responds to spatial frequencies of edges of white lines; replacing a set value of a pixel with 0 if an absolute value of a filter response corresponding to the pixel is greater than or equal to a second threshold value and the set value of the pixel is 1; and extracting a line segment vector estimated to correspond to a crack by using a transform that estimates continuity of the plurality of pixels.