Noise filtering is a crucial task in digital image processing, performing the function of preprocessing. In this paper, we propose an algorithm that employs deep convolution and soft thresholding iterative algorithms to extract and learn the features of noisy images. The extracted features are acquired through prior and sparse representation theory for image reconstruction. Effective separation of the image and noise is achieved using an end-to-end network of dilated convolution and fully connected layers. Several experiments were performed on public images subject to various levels of Gaussian noise, in order to evaluate the effectiveness of the proposed approach. The results indicated that our algorithm achieved a high peak signal-to-noise ratio (PSNR) and significantly improved the visual effects of the images. Our study supports the effectiveness of our approach and substantiates its potential to be applied to a broad spectrum of image processing tasks.

In the field of image denoising, convolutional neural networks (CNNs) have become increasingly popular due to their ability to learn effective feature representations from large amounts of data. In the field of image denoising, CNNs are widely used to improve performance. However, increasing network depth can weaken the influence of shallow layers on deep layers, especially for complex denoising tasks such as real denoising and blind denoising, where conventional networks fail to achieve high-quality results. To address this issue, this paper proposes a hybrid dilated convolution-based denoising network (AMDNet) that incorporates attention mechanisms. In specific, AMDNet consists of four modules: the sparse module (SM), the feature fusion module (FFM), the attention guidance module (AGM), and the image residual module (IRM). The SM employs hybrid dilated convolution to extract local features, while the FFM is used to integrate global and local features. The AGM accurately extracts noise information hidden in complex backgrounds. Finally, the IRM reconstructs images in a residual manner to obtain high-quality results after denoising. AMDNet has the following features: (1) The sparse mechanism in hybrid dilated convolution enables better extraction of local features, enhancing the network’s ability to capture noise information. (2) The feature fusion module, through long-range connections, fully integrates global and local features, improving the performance of the model; (3) the attention module is ingeniously designed to precisely extract features in complex backgrounds. The experimental results demonstrate that AMDNet achieves outstanding performance on three tasks (Gaussian noise, real noise, and blind denoising).

Obtaining the surface temperature of billets in heating furnaces has been a hot research in metallurgical industry applications. In order to accurately identify the billet location in infrared images and thus obtain the surface temperature of billets, this paper proposes a real-time segmentation network model based on multi-scale feature fusion to solve the problems of low resolution, low accuracy and slow detection speed of infrared images of traditional target image detection methods. In our method, a dataset with billet infrared images as the experimental object is firstly established, and the proposed network structure adopts multi-scale feature fusion to enhance the information interaction between feature maps at all levels and reduce the information loss during up-sampling by a dense up-sampling strategy. Meanwhile, a lightweight backbone network and deep separable convolution are used to reduce the number of network parameters and speed up the network inference, finally realizing real-time and accurate segmentation of the infrared images of blanks. The highest accuracy of the model in this paper reaches 94.89 $$\%$$% . Meanwhile, an inference speed of 80fps is achieved on GTX2080Ti. Compared with the existing mainstream methods, the method in this paper can better meet the real-time and accuracy requirements of industrial production.

The microstructure of superalloy materials has a decisive impact on its service performance. When preparing the material and photographing the microstructure, different depths of metallography perpendicular to the cut plane appear in the microstructure image. These metallographic features are major factors causing inaccurate segmentation. Aiming at the problems of traditional image processing methods, such as large noise and poor robustness of the edge extraction, the deep learning method is introduced. The receptive field of the traditional convolutional neural network method is too local to obtain remote dependencies, and dense feature extraction also brings problems such as excessive noise. To address the problem of non-tangential metallographic information in microstructure images that is not easily distinguishable. We consider how to combine and utilize the feature maps of the intermediate process as effectively as possible, and find that reusing affinity matrices from different angles and stacking them can improve the overall effect, and can solve the problem of inaccurate segmentation of metallographic images at different depths. We propose to improve and optimize the non-local attention module and further combine the module with the UNet network to form a new improved SNL-Unet image segmentation structure, which significantly raises the accuracy and efficiency of image segmentation, Additionally, we measure the characteristic parameters such as volume fraction, average thickness and the degree of rafting. The code for this paper will be available at github.com/ustbjdl1021/improved-snl-unet.

In recent years, human pose estimation, as a subfield of computer vision and artificial intelligence, has achieved significant performance improvements due to its wide applications in human-computer interaction, virtual reality, and smart security. However, most existing methods are designed for single-person scenes and suffer from low accuracy and long inference time in multi-person scenes. To address this issue, increasing attention has been paid to developing methods for multi-person pose estimation, such as utilizing Partial Affinity Field (PAF)-based bottom-up methods to estimate 2D poses of multiple people. In this study, we propose a method that addresses the problems of low network accuracy and poor estimation of flexible joints. This method introduces the attention mechanism into the network and utilizes the joint point extraction method based on hard example mining. Integrating the attention mechanism into the network improves its overall performance. In contrast, the joint point extraction method improves the localization accuracy of the flexible joints of the network without increasing the complexity. Experimental results demonstrate that our proposed method significantly improves the accuracy of 2D human pose estimation. Our network achieved a notably elevated Average Precision (AP) score of 60.0 and outperformed competing methods on the standard benchmark COCO test dataset, signifying its exceptional performance.