关键词: 手势识别, 卷积神经网络, CMOS图像传感器, 读出电路

Abstract: Hand gesture recognition technology has garnered significant attention in the fields of human-computer interaction and virtual reality. However, vision-based gesture recognition systems are still limited by chip area and power consumption. A readout circuit suitable for the sliding convolution operation of a 3×3 convolution kernel is proposed. It includes convolution and rectified linear unit-max pooling (ReLU-MaxPool) dual-function circuits. The convolution circuit can achieve configurable positive and negative weights, utilizing current weighting to perform high-precision convolution operations in the current domain. The ReLU-MaxPool dual-function circuit can simultaneously achieve ReLU and MaxPool functions. It employs switch-capacitors, registers, and customized logic circuits, and combines timing design to greatly improve work efficiency and operating speed. The proposed circuit is implemented based on 55 nm CMOS process with a supply voltage of 2.5 V. The output error of the convolution circuit under different process corners is less than 0.25%. The convolutional circuit is simulated using Monte Carlo simulation, with a mean error of 0.05% and a variance of 1.26%. By modeling the circuit errors and incorporating them into the gesture recognition algorithm, the accuracy on the test set decreases from the ideal algorithm model's 91.66% to 90.62%, with an impact of only 1.04 percentage points on the results.

Key words: hand gesture recognition, convolutional neural network, CMOS image sensor, readout circuit