Efficient Line Detection Method Based on 2d Convolution Filter

Abstract

The article proposes an efficient line detection method using a 2D convolution filter. The proposed method was compared with the Hough transform, the most popular method of straight lines detection. The developed method is suitable for local detection of straight lines with a slope from -45˚ to 45˚. Also, it can be used for curve detection which shape is approximated with the short straight sections. The new method is characterized by a constant computational cost regardless of the number of set pixels. The convolution is performed using the logical conjunction and sum operations. Moreover, design of the developed filter and the method of filtration allows for parallelization. Due to constant computation cost, the new method is suitable for implementation in the hardware structure of real-time image processing systems.

Pyramidal method of generalized spatially connected processing and an example of its implementation in image processing

“…Execution of relation (2) leads to the fact that when changing the orientation of the object, such as its rotation, the energy spectrum of its constituent elements with a given approximation δ remains constant. According to the proposed model of the form (1), the essence of the method of generalized spatially-connected preparation is as follows 21,22 . The m delayed signals are formed by delaying the video signal for a τ i time for the delayed signal numbered i, where i = 1,m.…”

Inverse correlation filters of objects features with optimized regularization for image processing

“…The task is to expand filter action on entire set of object fragments with some allowable error of the object recognition. The problem solution is found by introduction to (1) constraints on the feature vector h in the manner of regularization 17,18,19 .…”