EDGE DETECTION USING SPECTRUM ANALYSIS FOR LOW POWER ANALOG SMART CAMERA SENSOR

Abstract

This work presents an intelligent analog image sensor system for smart camera with a sliding discrete Fourier transform (SDFT) kernel. A complex-valued kernel based on the SDFT has been introduced for edge detection operation. These kernels are implemented in the form of resistors, which convolves with incoming raw data from the CMOS image sensor. The SDFT-based based kernel performs better in terms of egde connectivity, removal of irrelevant edges, and noise. The system consists of a 3x3 read-out CMOS image sensor, an edge detection kernel and additional circuitry like operational amplifiers and comparators to compute a 1bit image with the edges present in the taken photo. This framework has been implemented and tested using LTspice for circuit simulations and integrated with Matlab for performance evaluation of the outputs. The proposed SDFT approach outperforms Prewitt and Sobel methods interms of accuracy. Near-sensor analog processing approach has reduced the data size for transmission and storage by 90%. Furthermore, this approach does not require the usage of an analog-to-digital converter (ADC) during the processing. The near-sensor analog processing circuit consumes 200 mW power on average for the tested image data.