Automatic insect identification method based on deep learning

• Main Authors: WANG XIAOHUA, LI JUAN, PAN XIAOGUANG, SONG XIAOCHEN, JIAO LULU
• Format: Patent
• Language: Chinese; English
• Published: 05.07.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention belongs to the technical field of machine learning, and particularly relates to an automatic insect identification method based on deep learning, which comprises the following steps: data acquisition: acquiring a real tiger beetle image data set, marking the category of the real tiger beetle image data set, and completing the construction of a data set required by model training; the color intensity distribution of an original data set is not uniform, the size of a protruding part of each image is greatly changed, and the protruding part contains background noise, so that the recognition accuracy is not high, the original data set is correspondingly processed through preprocessing, the preprocessing comprises normalization, data cutting and data scale unification, and the recognition accuracy is improved; a data set is amplified, and the model training effect is ensured; construction: constructing a fusion classification model by adopting deep learning and machine learning related technologies,