VSIP

In this paper, we improve the natural scene text detection and recognition technology based on 2d attention and encoder-decoder framework. Firstly, the related work of text detection and recognition in different natural view is discussed. Secondly, we ...

The oracle bone script is the oldest writing system in China, and as the origin of Chinese characters, it is also the earliest cultural heritage in China. With the development of computer technology, deep learning has become an important technology of ...

In recent years, elevator safety accidents occur frequently, and relevant news has aroused people's strong attention to elevator safety. Among them, electric vehicle explosion accident hot spot is the highest. In this paper, according to the elevator ...

Aiming at the problems of low accuracy caused by low quality frames such as occlusion and noise, and the inability to effectively utilize local features of cross-channel in video surveillance, a video-based person re-identification method based on ...

The use of computer vision techniques to distinguish fresh from stale fruit has gained widespread acceptance in recent years. However, the scarcity of datasets in this field has made it more difficult for many of the research results to be accessible in ...

The progress made in deep learning has allowed the deployment of more powerful biometric authentication systems instead of traditional ones based on passwords or PIN codes. Facial recognition is widely used on smartphones to grant user access. However, ...

Personnel security has always been a hot topic of research especially in power systems. Effective detection of people in power scenes has become a high priority. Currently, algorithms for personnel detection have been developed considerably, and they ...

Change detection is a challenging problem in remote sensing applications. In recent years, many Convolutional Neural Network (CNN)-based change detection methods have been proposed due to the rapid development of deep learning techniques. First, the ...

Ship detection in complex weathers is an important application of object detection approaches in real life. In recent years, ship detection approaches for single weather have gradually matured. However, there are few ship detection approaches for ...

As a convenient biochemical detection technology, colloidal gold strip lateral flow immunochromatography has been widely used in disease detection and diagnosis, food safety and other fields. In order to achieve high accuracy and efficient test strip ...

Block-wise compressive sensing (BCS) uses a fix size sensing matrix to raster-scan the entire image to acquire the signal block by block. Compared with traditional image compressive sensing methods, BCS can greatly reduce the memory consumption of ...

Few studies present how to implement super resolution (SR) on abdominal Computed Tomography (CT) images. However, in the field of medical imaging, the development of an advanced SR system will potentially improve the clinical diagnosis and prognosis of ...

Building an effective fashion recommendation system is challenging due to the high level of subjectivity and the semantic complexity of the features involved. Users’ decision depends largely on their interest and the appearance of the product. Such ...

The Embodied AI is the current frontier direction in the field of AI and is regarded as a research leading to general artificial intelligence. Embodied AI refers to the study and development of artificial agents that can perceive, reason, and interact ...

Traditional no-reference image quality assessment algorithms mostly focus on the objective indicators of the image, such as the IE (information entropy) and clarity of the image. These indicators can reflect the quality of the image at the objective ...

Keypoint feature extraction algorithms based on manual feature rely on the professional domain knowledge of the designer, and usually perform well in depicting the detail of an image. However, these manual feature-based feature extraction algorithms ...

Hyperspectral images(HSIs) have rich spatial and spectral information, which provide a strong basis for distinguishing different landcover objects. The HSI classification methods based on deep learning have achieved good performance. Recently, several ...

Prior self-supervised video object segmentation models directly find pixel correspondences between pairs of frames. Instead, we propose a novel approach of employing superpixel features for learning visual correspondences between frames of a video ...

Existing video summarization methods are classified into either shot-level or frame-level methods, which are individually used in a general way. This paper investigates the underlying complementarity between the frame-level and shot-level methods, and a ...

Deep learning (DL) makes radar-based human activity recognition (HAR) attract more attention in the fields of intelligent security, traffic management, medical rehabilitation, and military operation, it can automatically extract the human activity ...

Aiming at the limitations of the prior art in the sorting of frequency hopping signals under the condition of low signal-to-noise ratio, this paper proposes a method to reduce the noise of the image through image preprocessing, and then improve ...

The conventional monopulse method to estimate target direction of arrival (DOA) performs well when the power of target is high, but its performance degrades seriously in strong clutter environment when the target power is lower than the clutter power. ...

Continuous Phase Modulation (CPM) has been widely used in satellite communication and mobile communication systems due to its good constant envelope characteristics and high bandwidth and power utilization. CPM signals can be divided into Single-h CPM ...

Non-negative Matrix Factorization (NMF) solves the single-channel speech separation problem. However, problem solving mostly separates the pure speech signal from the mixed speech signal with noise. This paper proposes an NMF algorithm based on ...

In this paper, an effective direction of arrival (DOA) estimation algorithm of coherent signals for cylindrical conformal arrays (CCAs) based on fourth-order cumulation is proposed. Affected by the shadow effect, the CCAs is divided into several ...