A Novel Adaptive Local Dimming Backlight Control Chip Design Based on Gaussian Distribution for Liquid Crystal Displays

Abstract

In this paper, a novel hardware-oriented adaptive local dimming backlight control algorithm is proposed for very large-scale integration implementation. The proposed algorithm includes a novel dimming algorithm to save energy of the liquid crystal display (LCD) backlight module and a novel contrast enhancement algorithm to compensate the dimming of brightness in displaying images. In order to improve the power-saving rates and image quality, a block-based model and a Gaussian distribution model were added. After obtaining a clipped point by the Gaussian distribution model, a pixel compensator was used to compensate the brightness value of each pixel to satisfy the human vision. In the meantime, a power width modulation module produced power control signal for the LCD backlight module. To meet the performance standards for displaying video in real time, the proposed algorithm was implemented by an eight-stage pipelined architecture. In addition, a new color-space YEF transform composed of integer adders and shifters was used to realize the grayscale converter to replace floating point multipliers. Moreover, the high-complexity operations, such as square root, multiplication, and division, required for the Gaussian distribution model were simplified by using a lookup table technique, which could reduce the hardware cost of the proposed design efficiently. This design was fabricated by a TSMC 0.18-μm CMOS process, and the chip area is 630 105 μm ² . Compared with previous local dimming backlight control algorithms, this work improved the average peak signal to noise ratio by 6.89 dB and saved 7.61% more backlight power consumption than previous studies.