Background: Harris corner detection extracting corner features based on the characteristic
value of the second order matrix, is regarded as one of the most successful algorithms in corner
Methods: OTSU algorithm abbreviation of the maximum variance between clusters can calculate the
maximum variance to distinguish the background area and the target area based on the image gray
histograms. To obtain the adaptive threshold, Harris needs to artificially select threshold and threshold
estimation becomes very difficult.
Results: To solve these problems, OHO algorithm is proposed in this paper which aims to optimize the
Harris algorithm based on OTSU. The OHO algorithm combines the characteristics of the high accuracy
of Harris algorithm and the adaptive threshold selection of OTSU.
Conclusion: Experiments show that the OHO algorithm can detect more details and authentic corners,
and has better adaptability and robustness than traditional Harris.