Depression is one of the most common mood disorders and the number of patients increases significantly in recent years. Due to the lack of biomarkers, conversation between patients and psychiatrists is still the main clinical diagnostic method which is easily influenced by subjectivity of both patients and psychiatrists. Synthetic House-tree-person test (S-HTP), a convenient and efficient mental assessment tool, minimizes subjective influences from patients, while its effectiveness is limited by the professional ability of analyst. Here we introduce a deep learning model DeHTP, a flexible and convenient depression detection method based on S-HTP without interaction between people. Experimental results demonstrate that DeHTP achieves 0.963 AUC and 0.9 accuracy, and outperforms the conventional manual analysis of S-HTP, which is conducted on the guideline of 50 conclusions from previous study related to depression. In addition, it reveals 22 depression-correlated drawing features aligned with conclusions above from the perspective of our proposed model. Leveraging the advantages of deep learning and S-HTP, this approach has the potential for widespread promotion and adoption as the available tool for daily self-mental monitoring, as well as the promising auxiliary diagnostic method in clinical.

Deep learning methods in computer vision have shown tremendous progress recently. Most image restoration methods are based on single feature extraction. However, image restoration requires various types of information to reconstruct the properties such as structure, color, texture details. Low-quality pictures are widespread in the real world because of the natural conditions and unstable photography. Therefore, we propose a double feature fusion network (DFFNet) to overcome this challenge in image inpainting. DFFNet extracts deblurring and structure features simultaneously and adopts multipath refinement framework aided by multipath contextual attention modules to restore the mask region in a coarse-to-fine manner. Thus, our work can restore blurred and large masked images into sharp and complete images and neglect the meaningless occlusions. It outperforms the state-of-the-art combined methods for sharper and smarter inpainting. First, we design a multi-path refinement network that can extract multi-features for further refinement and coarse inpainting. Second, multipath contextual attention modules are reconstructed for receiving edge and deblur features and finer inpainting results. Third, multi-stage synthesis loss function and double feature fusion units ensure that the restored images have the same structure and texture similarity as the original ground-truth counterparts. Therefore, the progressive learning framework DFFNet adopts key elements of image restoration information such as structure and texture, and refines the background for removing meaningless noise for sharper and smarter inpainting.

With the upgrading of the high‐performance image processing platform and visual internet of things sensors, VIOT is widely used in intelligent transportation, autopilot, military reconnaissance, public safety, and other fields. However, the outdoor visual internet of things system is very sensitive to the weather and unbalanced scale of latent object. The performance of supervised learning is often limited by the disturbance of abnormal data. It is difficult to collect all classes from limited historical instances. Therefore, in terms of the anomaly detection images, fast and accurate artificial intelligence‐based object detection technology has become a research hot spot in the field of intelligent vision internet of things. To this end, we propose an efficient and accurate deep learning framework for real‐time and dense object detection in VIOT named the Edge Attention‐wise Convolutional Neural Network (EAWNet) with three main features. First, it can identify remote aerial and daily scenery objects fast and accurately in terms of an unbalanced category. Second, edge prior and rotated anchor are adopted to enhance the efficiency of detection in edge computing internet. Third, our EAWNet network uses an edge sensing object structure, makes full use of an attention mechanism to dynamically screen different kinds of objects, and performs target recognition on multiple scales. The edge recovery effect and target detection performance for long‐distance aerial objects were significantly improved. We explore the efficiency of various architectures and fine tune the training process using various backbone and data enhancement strategies to increase the variety of the training data and overcome the size limitation of input images. Extensive experiments and comprehensive evaluation on COCO and large‐scale DOTA datasets proved the effectiveness of this framework that achieved the most advanced performance in real‐time VIOT object detection.

Image deblurring is a classic and important problem in industrial fields, such as aviation photo restoration, object recognition in robotics, and autonomous vehicles. Blurry images in real‐world scenarios consist of mixed blurring types, such as a natural motion blurring owing to shaking of the camera. Fast deblurring does not deblur the entire image because it is not the best option. Considering the computational costs, it is also better to have an alternative kernel to deblur different objects at a high‐semantic level. To achieve better image restoration quality, it is also beneficial to combine the blurring category location and important structural information in terms of specific artifacts and degree of blurring. The goal of blind image deblurring is to restore sharpness from the unknown blurring kernel of an image. Recent deblurring methods tend to reconstruct prior knowledge, neglecting the influence of blur estimation and visual fidelity on image details and structure. Generative adversarial networks(GANs) have recently been attracting considerable attention from both academia and industry because GAN can perfectly generate new data with the same statistics as the training set. Therefore, this study proposes a generative neural architecture and an edge attention algorithm developed to restore vivid multimedia patches. Joint edge generation and image restoration techniques are designed to solve the low‐level multimedia retrieval. This multipath refinement fusion network (MRFNet) can not only perform deblurring of images directly but also individual the frames separately from videos. Ablation experiments validate that our generative adversarial network MRFNet performs better in joint training than in multimodel. Compared to other GAN methods, our two‐phase method exhibited state‐of‐the‐art performance in terms of speed and accuracy as well as has a significant visual improvement.

A foreground segmentation method, including image enhancement, trajectory classification and object segmentation, is proposed for moving cameras under low illumination conditions. Gradient-field-based image enhancement is designed to enhance low-contrast images. On the basis of the dense point trajectories obtained in long frames sequences, a simple and effective clustering algorithm is designed to classify foreground and background trajectories. By combining trajectory points and a marker-controlled watershed algorithm, a new type of foreground labeling algorithm is proposed to effectively reduce computing costs and improve edge-preserving performance. Experimental results demonstrate the promising performance of the proposed approach compared with other competing methods.