Local Dimming Algorithm For Direct-lit Type LED Backlight In LCD TVs

Nowadays Liquid Crystal Display (LCD) has become the mainstream product in the field of flat panel display. As the fact that Liquid crystal can not emit lights itself, the backlight is necessary to provide light to illuminate the display area. With the advantages of green environmental protection, high color purity, wide color gamut, long life and fast response time, Light Emitting Diode (LED) has occupied the almost entire backlight market. The backlight modules of LCD could be side-emitting type or direct-emitting type. The uniformity and brightness of the side-emitting backlight module are not very well when the liquid crystal display size is large, because the light guide plate is greatly more complexity than ever. In contrast, the direct-emitting backlight module is quite simple and does not need the light guide plate, guaranteeing high brightness and uniformity of LCD. With the development of the high-power LED backlight lens in recent years, the direct-emitting backlight presents the merits on cost and structure.The Liquid crystal display has the inherent problem of low contrast, and under the ordinary conditions, for most nowadays TV, the backlight source always works in the highest luminance, causing power waste. Dynamic dimming algorithm can change the luminance of backlight source simultaneously with the display image content, in order to save the power energy and to enhance contrast. In this thesis, through researching on the dynamic dimming algorithms, a new local dimming algorithm is proposed based on image classification and is applied on the direct type high-power LED backlight of LCD TV. Through the design of backlight lens, each LED can control a single image block. The process of LED backlight diffusion simulation model is established and the light diffusion function template parameters are determined by measurement, which can obtain the adjustment factors of each liquid crystal pixel accurately and improve the compensation accuracy of liquid crystal pixels. The simulations are performed on both static pictures and dynamic videos, showing that the algorithm is validated and optimized. The principle prototype is developed and tested. The test results show that with the algorithm proposed in this paper on the principle prototype, the average energy saving reaches 34.41% and the average contrast enhanced 117.13% of 50 static images. The LCD TV prototype with the proposed dynamic local algorithm has advantages of simple structure, low cost, energy saving and environmental protection.