An IGA Model for Resource Allocation in Heterogeneous Networks for the Internet of Things

Dynamic obstructions such as trees in the road environment can affect the quality of images captured by industrial cameras. In order to improve the quality of image acquisition in high dynamic scenes, this study proposes a new imaging algorithm based on convolutional neural networks and combined with single exposure high dynamic range imaging methods. This algorithm is mainly divided into a color camera image signal processing link module and a single exposure high dynamic range image enhancement module. The above two modules can improve the resolution of high dynamic range images. The experimental results show that the algorithm proposed in this study can achieve a response time of 63ms at the fastest, with PSNR, SSIM, and perceptual loss scores reaching 38.67, 0.976, and 0.087, respectively. Research has found that this algorithm has good application value in the field of high dynamic range image imaging.