为增强变电站中电力设备红外热像图的视觉效果,突出局部热异常区域,方便故障定位及检修,提出了一种基于自适应遗传算法的红外热像图模糊增强技术。对小波变换后的红外子热像图,进行同态滤波增强处理和模糊增强处理,并运用动态自适应遗传算法对模糊参数进行优化,最后,经小波重构得到了效果增强的红外图像。实验结果表明,该方法相对同态滤波、模糊集增强、遗传模糊等算法,红外图像的边缘强度、对比度、清晰度分别至少提高12.6%、27.7%、33.7%,有利于检修人员进行电力设备的热异常定位及故障诊断。
In order to enhance the visual effect of infrared image of electric power equipment in substation, to highlight the thermal anomaly area, and to help engineers analyze the faults, a fuzzy enhancement technology infrared image based on the adaptive genetic algorithm was proposed. After the wavelet transform of the infrared image, homomorphic filtering enhancement and fuzzy enhancement were processed, and the dynamic adaptive genetic algorithm was used to optimize the parameters of fuzzy method, finally, the images were reconstructed. Experimental results show that the effect of proposed method is better than enhancement homomorphic filtering, fuzzy enhancement, and fuzzy genetic algorithm, and the image contrast, resolution, and clarity can be improved by at least 12.6%, 27.7%, and 33.7% respectively. It is favorable for the maintenance of electric power equipment, especially in thermal anomaly location and fault diagnosis.