Image Processing

This dataset selects the CIFAR-10 small-scale general object recognition dataset, which contains a total of 60,000 RGB color images with a size of 32×32, belonging to 10 categories. The dataset is divided into a training set of 40,000 images, a testing set of 10,000 images, and a validation set of 10,000 images. The images in the dataset are processed with bit masking, and the image classification results are compared with those of the original dataset, providing an experimental basis for the study of how bit masking improves the accuracy and speed of the algorithm.

‌DRIVE Dataset‌: The DRIVE dataset consists of 40 fundus images collected from 7 patients with retinal disease and 33 healthy individuals. Each image has a resolution of 584 × 565 pixels. A total of 20 images are used for training, while the remaining 20 images are designated for testing. Each image has two annotations provided by two different experts, and we use the first annotation as the ground truth.

This paper proposed a PE-VAE-GAN network that adaptively selected image reconstruction networks based on flow pattern classification, significantly improving the quality of Electrical Resistance Tomography (ERT) reconstructed images.To address insufficient feature extraction from voltage data,we presented a pseudo-image encoding method that converted the one-dimensional voltage signals into the two-dimensional grayscale images.

The system consists of an UAVs remote sensing system and an edge computing system. The core components of the UAVs remote sensing system mainly consist of the imaging system K510 development board and a 5G module. The K510 comes equipped with a camera and an LCD screen. The edge computing system constructed in this paper utilizes the NVIDIA Jetson series development kit, which comes with a GPU module to enhance digital image processing capabilities. The captured raw images are stitched and displayed on the NVIDIA Jetson edge computing platform using our designed improved SFIT algorithm.