基于显微光学成像的微细纤维测量识别系统光源优化研究

显微光学成像的纤维识别系统的照明光源,对成像质量有很大影响。由于单个LED光源发光功率有限,且在照射面的光分布不均,导致系统采集到的微纳米级纤维(单丝平均直径15~40μm)图像模糊,影响测量与识别结果的准确性。根据LED的光学特性,对系统的照明光源进行优化设计,设计了3种不同的LED平面光源阵列,并根据LED阵列光源在照射面上的照度表达式,仿真模拟给出LED阵列的照度分布,并以实验进行验证,对系统实际检测的图像进行图像识别处理和图像参数峰值信噪比评价,给出优化结果。研究结果表明,优化后的LED平面光源阵列有效地解决了照度不均导致的微细纤维图像模糊识别不清的问题,优化后得到的纤维样品图像边缘清晰,有效提高了系统检测纤维样品的图像质量,为微细纤维测量识别与研究提供了新的途径。

Research on Light Source Optimization of Microfiber Measurement and Recognition System Based on Micro-optical Imaging

The illumination source of the microscopic optical imaging fiber recognition system has a great influence on the imaging quality. Due to the limited luminous power of a single LED light source and uneven light distribution on the illuminated surface, the micro-nano fiber (average diameter of monofilament 15~40μm) collected by the system was blurred, which affected the accuracy of measurement and recognition results. According to the optical characteristics of LED, the illumination source of the system was optimized. Three different LED planar light source arrays were designed. According to the expression of the illuminance of the LED array on the illuminated surface, the simulation showed the LED array. The illuminance distribution was verified by experiments. The image recognition process and the peak signal-to-noise ratio (PSNR) of the image parameters were evaluated for the images actually detected by the system, and the optimization results were given. The research results showed that the optimized LED planar light source array effectively solved the problem of unclear blur recognition of microfibril images caused by uneven illumination. The image of the fiber sample obtained after optimization was clear, which effectively improved the image quality of the system for detecting fiber samples, and this provided a new way for microfiber measurement identification and research.