Method for creating rice plant deep learning counting data set based on CECI algorithm

• Main Authors: GU SUSONG, CHEN SHIQI, BAI XIAODONG, ZHAO LAIDING, XIE TINGTING, JI ZHONGYUAN, LIU PICHAO
• Format: Patent
• Language: Chinese; English
• Published: 11.11.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a method for creating a rice plant deep learning counting data set based on a CECI algorithm, and the method comprises the steps: collecting original rice picture data, carrying out the vegetation chromaticity screening and HSV space chromaticity segmentation processing of an original rice picture, and extracting a green rice region in the processed original rice picture; performing chromaticity enhancement on the original rice picture, and performing ten-cluster clustering on the original rice picture after chromaticity enhancement processing; calculating the coincidence rate of the green rice area and the ten-cluster clustering block, and combining and splicing according to the coincidence rate to obtain a final rice picture; and performing corrosion expansion processing on the final rice picture, calculating a connected domain of the rice, correcting the picture generated by computer automatic labeling according to the barycentric coordinates of the connected domain, and creating a