WiMi Hologram Cloud Inc (WIMI)

[StockpickerCAPC]

WiMi Develops Convolutional Neural Network-Based Image Fusion Algorithm System.

Source
https://www.newstrail.com/wimi-develops-convolutional-neural-network-based-image-fusion-algorithm-system/

December 19, 2022

In recent years, image fusion has become a research hotspot in image understanding, computer vision, and remote sensing. It is widely used in medicine, intelligent robotics, industry, and other areas, and the study of image fusion algorithms has important theoretical significance and practical value. Meanwhile, with the continuous improvement of the computer level, the theory of convolutional neural networks is also developing rapidly, and it has been widely used in many fields, such as target recognition and face recognition. The application of convolutional neural networks to image fusion has obvious superiority, which can improve the feature extraction and assignment in image fusion and enhance the quality of fused images.

It is understood that WiMi Hologram Cloud, Inc. (NASDAQ:WIMI) has developed a convolutional neural network-based image fusion algorithm system. Image fusion is to process and fuse two or more images acquired by different sensors to achieve complementary information between images, maximize image quality, generate content-rich merged images that better match visual perception, and then complete the analysis and processing of information. The convolutional neural network is a typical deep learning model that learns feature representation mechanisms at different levels of abstraction for signal or image data. CNN extracts features of the input image by learning filters to obtain other feature maps at each level, and each unit or coefficient in the feature map is called a neuron. Different computational methods, convolution, activation function, and pooling, are generally utilized to connect the feature maps between adjacent levels.

The key to WiMi’s convolutional neural network-based image fusion algorithm is to effectively extract image information and use an appropriate fusion strategy to maximize the useful information from the source image and fuse it into the resultant image to obtain a high-quality image.

Firstly, the images to be fused are acquired and preprocessed. The preprocessed images are input to the convolutional neural network separately for training to extract their image fusion features. Then the optimal thresholding method is used to segment the fusion features and fuse different regions of different images accordingly to obtain the final image fusion results.

A convolutional neural network is a multilayer structure, including an input layer, pooling layer, fully connected layer, convolutional layer, etc. The convolutional layer is the most critical part, which contains several neural network nodes that can extract the features for image fusion. The pooling layer can downscale the image fusion features, obtain the new image fusion feature mapping set, and then iterate continuously through the weights for training and learning. To perform the optimal fusion of images, the images must be segmented into different regions. The images to be fused are divided into other areas by the optimal segmentation threshold, and the various parts are merged to output the image fusion results.

The image fusion algorithm system based on the convolutional neural network of WiMi not only improves the clarity and brightness significantly but also improves the image signal-to-noise ratio and the image quality, which can obtain better image visual effect and has obvious superiority compared with the traditional image fusion technology.

WiMi’s convolutional neural network-based image fusion algorithm has become a critical image analysis and computer vision technology; with the continuous improvement of hardware and related research, its application in various fields will continue to develop further. WiMi has a wide range of applications in a target recognition, intelligent robotics, medical image processing, industrial Internet, etc.