Agricultural pest intelligent detection method

• Main Authors: CAI WEIMING, PANG HAITONG, MA LONGHUA, SU HONGYE
• Format: Patent
• Language: Chinese; English
• Published: 03.12.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention relates to an agricultural pest intelligent detection method. The method comprises the following steps: collecting agricultural pest images by adopting a data crawler and real-time image shooting mode; cleaning web crawler data in a manual selection mode, removing non-agricultural pest images and overlapped agricultural pest images, and the performing format normalization processing on selected crawler image data and real-time acquired image data in a script operation mode; processing the normalized image data by adopting a data enhancement means; marking all processed picture data by adopting an open source tool LabelImg to form an agricultural pest image data set; establishing an agricultural pest target detection model; training the agricultural pest target detection model; and performing precision evaluation on the trained agricultural pest target detection model. According to the method, the pest detection precision is high, the detection efficiency is also high, the labor intensity can be