Study Of Color Blindness Correction Method Based On Image Color Transform

Color blindness is the common disease of color vision deficiency(CVD),mainly caused by the descendiblity.The achromats can't discriminate some colors or identify colors,which brings much inconvenience in their life,studying,working and so on. To decrease the inconvenience and improve the ability of color discrimination for achromats,some researches are studied in many fields such as medicine,psychology and gene.However,it is noninvasive and more feasible to study CVD and re-coloring images for achromats from the image processingIn this dissertation by studying CVD and re-coloring images for achromats from the image processing,one simulating math model of anomalous trichromats and two methods of re-coloring images for achromats have been proposed,and then a testing system for achromats has been designed.Up to now the research on the model of anomalous trichromats mainly focuses on the qualitative analysis.In this dissertation,based on a simple dichromat model of Wachtler T,one simulating math model of anomalous trichromats has been proposed, which is a quantitative and intuitive simulating model.Experimental results show that the model of anomalous trichromats proposed by us can describe the perceptive change of color vision from dichromats to normal color vision people.To improve the ability of capturing information from color images for achromats, two methods,which are re-coloring images for dichromats based on an adaptive mapping algorithm and re-coloring images for achromats by rotating H component, have been proposed.Re-coloring images for dichromats based on the adaptive mapping algorithm is only suitable for dichromats.The method goes as follows, firstly it divides the color plane and sets up a mapping relation among colors according to the distribution of colors of an image,and then re-colors it based on the mapping relation among colors.In this dissertation,two methods based on the adaptive mapping algorithm are proposed according to the difference between mapping methods,which are the adaptive mapping algorithmⅠand the adaptive mapping algorithmⅡ.The former one is a nonlinear mapping method and is only suitable for testing image for achromats,while the latter is a linear one and is suitable for both testing images for achromats and natural images.Another method of re-coloring images for achromats by rotating H component is both suitable for dichromats and anomalous trichromats.In this method,the H component is rotated according to characteristics of its continuity and periodicity so that the ability of discriminating color is improved.In the computer simulation of testing images for achromats,the above three methods all have the good effect so that the information which can not be got from original images can be read from re-colored images by dichromats.Among these three methods,the adapting mapping algorithmⅠhas the best effect,while the method by rotating H component ranks the second,and the adapting mapping algorithmⅡranks the last.In the experiment of natural images for achromats,the best one is the method by rotating H component,and the second is the adapting mapping algorithmⅡ,while the effect of adapting mapping algorithmⅠis rather poor.In the dissertation,one testing system for achromats has also been designed.The system is divided into two parts,the testing system of images and the testing system of color balls.The testing system of images is to test the style and degree of achromats,while at the same time,it can realize the above three methods.The testing system for color balls is mainly to propose one method for distinguishing the protanopia and deuteranopia based on the model of dichromats.The system may provide more precise indices and convenience for testing achromats.