A Low Power Analog CMOS Vision Chip for Edge Detection Using Electronic Switches

Abstract

An analog CMOS vision chip for edge detection with power consumption below 20 mW was designed by adopting electronic switches. An electronic switch separates the edge detection circuit into two parts: one is a logarithmic compression photocircuit, and the other is a signal processing circuit for edge detection. The electronic switch controls the connection between the two circuits. When the electronic switch is off, it can intercept the current flow through the signal processing circuit and restrict the magnitude of the current flow below several hundred nA. The estimated power consumption of the chip, with 128 × 128 pixels, was below 20 mW. The vision chip was designed using 0.25 µm 1-poly 5-metal standard full custom CMOS process technology.

A CMOS vision chip for a contrast-enhanced image using a logarithmic APS and a switch-selective resistive network

“…8,9 The other uses two analog buffers to achieve raw and smoothed images. 7 These structures have their own disadvantages in power and area consumption, signal aliasing and additional noise. To overcome these problems, a switch-selective resistive network was applied.…”

“…The narrow output limits application of the bioinspired vision chip based on resistive network. 7,10 In this paper, we applied a metal-oxide-semiconductor (MOS)-type photodetector to a logarithmic APS to achieve a sufficient output swing for a bio-inspired vision chip in natural illumination condition.…”

“…Recently, bioinspired vision chips have been developed to overcome these problems. [6][7][8][9][10] The vision chips, which mimic the structure and functions of image capturing and processing in the human retina, offer several advantages including compact size, high speed, low power dissipation, and dense system integration.…”

“…[6][7][8][9][10]15 These additional circuits can cause additional area consumption, power dissipation and noise. In particular, the design of the structure, as well as the other circuits, is very important for improvement of their final output image quality.…”

“…1,2,[6][7][8][9][10] Photoreceptors, horizontal cells, and bipolar cells in human retina are related to extraction of edge from an incident image. 11 Photoreceptors transform the input light into electrical signals and yield the voltage in proportion to the input light intensity.…”