The power accessories have various types and models, and the storage and retrieval management with RFID technology cannot cover them all, which often leads to the inaccuracy and low efficiency of power accessories storage and retrieval, and the management quality not to meet the production requirements.In view of these problems, we carry out a research on intelligent recognition of power accessories based on machine learning and image recognition to correct the deficiency of RFID technology for storage and retrieval flexcon reverse osmosis water storage tank management of power accessories.Firstly, the gray-scale processing and binarization methods are used to process the original images, and the minimum circumscribed rectangle is used to correct the original images.Secondly, a deep 9x11 pergola learning model suitable for identifying power accessories is constructed using CNN and CRNN deep neural networks with combination of CTC loss function, and the suspected accessories are recommended synchronously according to the image recognition coincidence.
The images of power accessories are acquired by intelligent equipment, and their name and model are identified in real time using the proposed methods, with prompt of their overall dimension, application scope and product use.The experimental results show that the accuracy of the intelligent recognition of power accessories based on machine learning and image recognition reaches 95%, which significantly improves the intelligent level of warehousing management.