|
Fast, accurate and high-quality extraction of trace information is important in the fields of criminal investigation, archaeology and inspection. Reflection transformation imaging is based on a digital photography and surface reflection imaging model, and can record and process the surface texture information of the object, thereby realizing a three-dimensional visual rendering image of the object. In different occasions, the objects carrying the traces are different in the type and color, and the dimensions, types, colors, etc. of the traces are also different. In order to make the different trace features clear, this paper builds a detection device based on the principle of reflection transformation imaging. LED used in the device has four colors: white, red, green and blue. Traces with different colors and types have different reflectivity for different wavelengths of illumination light, thus a high contrast trace image is achieved by using the device. And the device can render the image and get clear visibility of traces. The local histogram equalization algorithm is further used to extract edge information of the traces of different rendering effects, and the detail information of the trace image is further enhanced. This paper introduces a non-reference image evaluation function to evaluate the quality of the image. In this paper, the device is used to detect the traces written by colorless pen, colorless ballpoint pen, black pen and black ballpoint pen on the white paper. Then the device is used to detect traces written by different color water-based pen and oily-based pen on the white paper. Compared with monochromatic source illumination, the device can obtain higher contrast trace information according to the color and type of the trace. Furthermore, edge extraction of these rendered trace images makes the traces easier to identify. For different colors of papers and traces, the device can obtain high contrast images by selecting the appropriate illumination wavelength. Therefore, the device will have important application prospects in the fields of criminal investigation and ancient cultural relics identification.
|
