The automatic recognition system for fruits and vegetables has gradually become an important means to improve agricultural production efficiency and reduce labor costs. A fruit and vegetable recognition method based on MobileNet-V2 is proposed, aiming to utilize the lightweight and efficient characteristics of the MobileNet-V2 model to achieve fast and accurate recognition of multiple fruits and vegetables. Firstly, MobileNet-V2 is used for pre training, with a focus on training its final layer. On this basis, an additional Flatten layer and four fully connected layers are added. The experimental results show that the accuracy of this method is 98.6% on the training set and 97.2% on the test set. It has high recognition accuracy and fast inference speed in fruit and vegetable recognition tasks.
Research on Fruit and Vegetable Recognition Method Based on MobileNet-V2
23.10.2024
529291 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Forecasting Fire Using MobileNet Architecture
IEEE | 2023
|Monitoring image anomaly detection based on lightweight network MobileNet
British Library Conference Proceedings | 2022
|Speeds fruit, vegetable ripening
Engineering Index Backfile | 1957
|RAILWAY TRACK FAULTS DETECTION BASED ON IMAGE PROCESSING USING MOBILENET
DOAJ | 2022
|