Image recognition processing method and device

Abstract

The invention provides an image recognition processing method and device. The method comprises steps that original image data, convolutional neural network configuration parameters and convolutional neural network operation parameters are acquired from a data transmission bus, the original image data comprises M pixel point data, and M is a positive integer; wherein the convolutional neural network operation module performs convolutional neural network operation on the original image data according to the convolutional neural network configuration parameters and the convolutional neural network operation parameters, and the convolutional neural network operation module comprises a convolutional operation unit, a batch processing operation unit and an activation operation unit which are connected in sequence. The embodiment of the invention further provides a corresponding device. According to the technical scheme provided by the embodiment of the invention, the real-time performance ofimage recognition processing is improved.

Description

technical field [0001] The invention belongs to the field of data processing, and in particular relates to an image recognition processing method and device. Background technique [0002] Convolutional Neural Network (CNN for short) was first proposed by Yann Lecun, and it was applied to handwritten digit recognition and has maintained its dominance in this field. In recent years, convolutional neural networks have continued to develop in multiple directions, and have made breakthroughs in speech recognition, face recognition, general object recognition, motion analysis, natural language processing, and even brain wave analysis. CNNs can be scaled up and configured to support labeling of datasets for learning processing. Under these conditions, CNNs have been found to be successful in learning complex and robust image features. [0003] A CNN is a feed-forward artificial neural network in which individual neurons are tiled in such a way that they respond to overlapping seg...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problem of poor real-time acquisition of parameters in the image recognition processing process of the prior art, the present invention provides an image recognition processing method and device, which realize real-time provision of convolutional neural network configuration parameters and operation parameters according to actual needs in the image recognition processing process

Images