Ship liquid pipeline leakage monitoring method based on machine vision

• Main Authors: DU TAILI, DUAN XUXU, LI SHUNQI, ZHANG PENG, ZHANG YUEWEN, WANG FEI, JIANG XINGJIA, ZOU YONGJIU, SUN PEITING
• Format: Patent
• Language: Chinese; English
• Published: 30.06.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a ship liquid pipeline leakage monitoring method based on machine vision, and the method is characterized in that the method comprises the steps: obtaining a ship liquid pipeline monitoring video image, and carrying out the preprocessing of the ship liquid pipeline monitoring video image, and generating a ship liquid pipeline image frame sequence; performing classification identification based on a KNN algorithm on the ship liquid pipeline image frame sequence, and judging whether each image frame has a leakage condition or not; carrying out foreground extraction on the image frame with the leakage condition by adopting a Gaussian mixture GMM background subtraction method to obtain leaked liquid drops; and performing leakage feature extraction on the extracted foreground image to position a leakage point. According to the invention, leakage detection is carried out on the ship pipeline monitoring video image by using an image processing technology, and efficient positioning and accurate