Traffic Light Recogintion Solution For The Color-vision Deficient Drivers

Timely and correctly identify the traffic signal not only affect the smooth running of traffic flow, but also relate to the raffic security problems.However, due to differences in the implementation and the accuracy of driver color vision test,a lot of anomalous trichromatism even color blind people into the motor vehicle driving field.In addition, the driver’s ability to identify traffic signals are also subject to a variety of factors, such as weather (night, snow, fog, dust and so on), personal status (fatigue, age, eye diseases, attention).Therefore, in order to assist the driver to identify traffic signals, improve road safety,this paper analyses the characteristics of the crowd abnormal color vision of color recognition proposed two kinds of road and vehicle auxiliary signal recognition program.On the road service signal identification scheme, for the color-vision deficient drivers, an economically feasible design of traffic lights is proposed to help them identify the signal under poor visual environment. Based on the generally accepted color-blind visible spectrum and a color-blind simulation software (Color Schemer Studio), the design takes account of both the characteristics of the existing traffic lights and of the visual impaired people, and generates low costs of physically rebuilding. The design achieved by install auxiliary lights with strong color contrast on traffic lights.The simulation results show the identifiability of signals in the design is clear enough, which would improve the traffic safety.In the scheme of vehicle auxiliary signal recognition,this paper proposes a automatic identification system of traffic signal based on computer image processing technology.At present, the application of video analysis technology in signal recognition is still a challenging research topic, has not yet formed a general algorithm can be used in most cases.According to the analysis of sample video, this paper proposes a traffic signal recognition method based on pixel statistics.In the HSV color space, by calculating the ratio of different pixels, to determine the current state of the signal lamp.Finally, on the basis of the above research, through the MATLAB software platform, this paper designed a simulation system, and verify the system using the sample videos. The experimental results show that the automatic identificationsystem of traffic signal has a certain precision.The results of the study for the standardized management of abnormal color vision driver has certain guiding value in china.