前置逐点亮度修正处理算法  

The preprocessing algorithm for point by point luminance correction

作　　者：郝亚茹[1] 邓招奇[1] 严飞[2] HAO Yaru DENG Zhaoqi YAN Fei(University of Electronic Science and Technology of China, Zhongshan Institute, Zhongshan 528402, China School of Information and Control, Nanjing University of Information Science and Technology, Nanjing 210044, China)

机构地区：[1]电子科技大学中山学院,广东中山528402 [2]南京信息工程大学信息与控制学院,江苏南京210044

出　　处：《中山大学学报（自然科学版）》2017年第2期80-87,共8页Acta Scientiarum Naturalium Universitatis Sunyatseni

基　　金：国家自然科学基金(61605083);中山市社会公益科技项目(2016B2145);电子科技大学中山学院青年基金项目(415YJ04);电子科技大学中山学院科研启动项目(415YKQ05)

摘　　要：逐点校正算法为基础,提出一种完全独立于显示控制系统的前置处理方案。该处理方案以DVI接口为传输介质,可完成对任意分辨率、任意格式的图像信息的亮度逐点校正。在现有逐点校正模型的基础上,实现前置处理方案中的亮度修正算法模型;根据算法模型中校正参量的聚集特性,同时采用压缩校正参量取值范围的方法和数值重组的方式来提高SDR SDRAM的频率,据此设计的数字电路可以实现对任意分辨率、任意传输格式的数据进行逐点亮度校正功能;通过理论和定量计算以及实际观测前置法和后置法的应用效果,得出结论:前置法产生的不一致性的灰度级数大于后者,并且灰度级越低,校正的不一致性越差,但前置法的灵活性和通用性明显优于后置法。Based on the brightness correction algorithm,a pre-processor scheme is proposed. The processor can realize the function of point by point correction for image data of arbitrary resolution and arbitrary transmission format. The pre-processor is completely independent of the display control system.Firstly,the model of point by point luminance correction algorithm is introduced. On the basis of the model,the model of point by point correction algorithm which is used to the pre-processor is derived.Then the design structure of the current hardware condition is analyzed based on this model,which mainly to SDRAM SDR frequency as the bottleneck analysis,According to the aggregation characteristics of the correction parameters,the method and numerical reconstruction of the range of compression correction parameters are adopted. The frequency of SDR SDRAM was indirectly improved. The data of any resolution and arbitrary transmission format can be handled by the point by point correction technology which is in the pre-processor. Finally,The application effect of pre method and post method is analyzed. It is concluded that the effect of pre method is worse than that of the post method,especially on the performance of low gray level uniformity. The flexibility and versatility of the prefix method is superior to the post method.

关 键 词：前置处理器 DVI传输格式 压缩取值 数值重组 

分 类 号：TN27[电子电信—物理电子学] TN702