The Research Of Realization Method And Technology Of Solar-blind Ultraviolet Image Fusion Based On I.MX6Q Processor

When the electric field intensity of the overhead transmission line surface exceeds the ionization strength of the air molecule,the self-sustaining discharge phenomenon,that is,corona discharge,occurs in the extremely uneven electric field.When the corona discharge phenomenon occurs in high voltage power equipment,it will be accompanied by sound,light,heat and other effects,causing interference to radio communication,so that the ozone and nitrogen oxides generated by air decomposition will corrode transmission lines and electrodes,resulting in power loss.Therefore,it is of great significance to accurately detect the occurrence of corona phenomenon and locate the occurrence position of corona early,so as to save energy resources and ensure the stable operation of power system.By analyzing the spectral characteristics of corona discharge,it is concluded that the method of "Solar-blind" ultraviolet detection with visible and ultraviolet double spectra has the characteristics of low false alarm rate,which is reliable and efficient.This paper mainly aims at the electronic part of the handheld double spectrum " Solar-blind " ultraviolet detector,and through the scheme demonstration,the i.MX6 Q embedded development platform with low power consumption and high performance of NXP company are selected for dual channel video image acquisition and fusion,which aimed at the completion of handheld corona mobile detection equipment technical verification.i.MX6 Q platform has a powerful video image processing subsystem,including IPU(Image Processing Unit),VPU(Video Processing Unit),GPU(Graphics Processing Unit),this paper constructs the driving framework based on the video image processing subsystem and V4L2.Through the IPU API to complete the acquisition and display of visible light channel 1920 ? 1080 HD video and ultraviolet channel 720 ? 576 SD video image,through the threshold processing algorithm based on OpenCL and GPU to complete the fusion operation of the double spectral image,and it completes the video and image file preservation and reading in the underlying hardware function.On the basis of kernel driver,the system running library layer and application layer of Android system are programmed,and the operation and display interface are designed.The final test results show that the human-computer interaction interface is friendly,the video acquisition image is stable and clear.The fused video frame rate is 20 frames/s,and the running effect is smooth,which lays a foundation for the subsequent research of ultraviolet &visible light two-channel detection.