Capsule endoscope image feature extraction method based on improved Harris angular points

• Main Authors: LIU TENGYUN, LIANG YANXING, WANG YINGHUI, ZHANG SHAOJIE
• Format: Patent
• Language: Chinese; English
• Published: 26.12.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a capsule endoscope image feature extraction method based on improved Harris angular points, and the method comprises the steps: calculating a partial derivative and a gradient value of each pixel point of a capsule endoscope image, and filtering the pixel points with the gradient values lower than a set gradient threshold value; a Harris corresponding function is improved, an improved Harrsi response function value of each pixel point is calculated, and potential feature points are obtained through the response function values; combining the improved Harris response function value with the window width and the sliding distance of a sliding window to obtain a self-adaptive sliding window, and performing non-maximum suppression on potential feature points; and determining the direction of each feature point, taking the direction as the main direction of the improved Harris corner, constructing an rBRIEF descriptor, and carrying out feature point matching by using the descriptor. Accordi