Improved lightweight identification of agricultural diseases based on MobileNetV3

07/19/2022

∙

by   Yuhang Jiang, et al.

∙

0

∙

share

At present, the identification of agricultural pests and diseases has the problem that the model is not lightweight enough and difficult to apply. Based on MobileNetV3, this paper introduces the Coordinate Attention block. The parameters of MobileNetV3-large are reduced by 22 by 19.7 MobileNetV3-small are reduced by 23.4 the accuracy is increased by 0.40 was migrated to Jetson Nano for testing. The accuracy increased by 2.48 98.31 deploying the agricultural pest identification model to embedded devices.