Automatic medical image segmentation has made great progress owing to the powerful deep representation learning. Inspired by success of self-attention mechanism in Transformer, considerable efforts are devoted designing robust variants encoder-decoder architecture with Transformer. However, patch division used existing Transformer-based models usually ignores pixel-level intrinsic structural features inside each patch. In this paper, we propose a novel framework called Dual Swin Transformer...

Palmprint is a unique and reliable biometric characteristic with high usability. With the increasing demand of highly accurate robust palmprint authentication system, multispectral imaging has been employed to acquire more discriminative information increase antispoof capability palmprint. This paper presents an online system that could meet requirement real-time application. A data acquisition device designed capture images under Blue, Green, Red, near-infrared (NIR) illuminations in less...

Palmprint processes a number of unique features for reliable personal recognition. However, different types palmprint images contain dominant features. Instead, only some the are visible in image, whereas other may not be notable. For example, low-resolution image has principal lines and wrinkles. By contrast, high-resolution contains clear ridge patterns minutiae points. In addition, three dimensional (3-D) possesses curvatures surface. So far, there is no work to summarize feature...

In this paper, a novel deep network is proposed for multi-focus image fusion, named Deep Regression Pair Learning (DRPL). contrast to existing fusion methods which divide the input into small patches and apply classifier judge whether patch in focus or not, DRPL directly converts whole binary mask without any operation, subsequently tackling difficulty of blur level estimation around focused/defocused boundary. Simultaneously, pair learning strategy, takes complementary source images as...

Accurate segmentation of organs or lesions from medical images is crucial for reliable diagnosis diseases and organ morphometry. In recent years, convolutional encoder-decoder solutions have achieved substantial progress in the field automatic image segmentation. Due to inherent bias convolution operations, prior models mainly focus on local visual cues formed by neighboring pixels, but fail fully model long-range contextual dependencies. this article, we propose a novel Transformer-based...

Graph and subspace clustering methods have become the mainstream of multi-view due to their promising performance. However, (1) since graph learn graphs directly from raw data, when data is distorted by noise outliers, performance may seriously decrease; (2) use a "two-step" strategy representation affinity matrix independently, thus fail explore high correlation. To address these issues, we propose novel method via learning <underline xmlns:mml="http://www.w3.org/1998/Math/MathML"...

Palmprint has proved to be one of the most unique and stable biometric characteristics. Almost all current palmprint recognition techniques capture 2-D image palm surface use it for feature extraction matching. Although can achieve high accuracy, images counterfeited easily much 3-D depth information is lost in imaging process. This paper explores a approach by exploiting structural surface. The structured light used acquire data, from which several types features, including mean curvature...

It has been reported that concentrations of several biomarkers in diabetics' breath show significant difference from those healthy people's breath. Concentrations some are also correlated with the blood glucose levels (BGLs) diabetics. Therefore, it is possible to screen for diabetes and predict BGLs by analyzing one's In this paper, we describe design a novel analysis system purpose. The uses carefully selected chemical sensors detect Common interferential factors, including humidity ratio...

Existing multi-label medical image learning tasks generally contain rich relationship information among pathologies such as label co-occurrence and interdependency, which is of great importance for assisting in clinical diagnosis can be represented the graph-structured data. However, most state-of-the-art works only focus on regression from input to binary labels, failing make full use valuable due complexity graph In this paper, we propose a novel framework based Graph Convolution Networks...

Learning discriminative features is of vital importance for automatic facial expression recognition (FER) in the wild. In this article, we propose a novel Slide-Patch and Whole-Face Attention model with SE blocks (SPWFA-SE), which jointly perceives locality characteristics informative global face effective FER. Specifically, well-designed slide patches are proposed to extract local features. Different from existing methods, our not only can maintain information at edge area patches, but also...

Accurate segmentation of organs or lesions from medical images is crucial for reliable diagnosis diseases and organ morphometry. In recent years, convolutional encoder-decoder solutions have achieved substantial progress in the field automatic image segmentation. Due to inherent bias convolution operations, prior models mainly focus on local visual cues formed by neighboring pixels, but fail fully model long-range contextual dependencies. this paper, we propose a novel Transformer-based...

Thanks for the cross-modal retrieval techniques, visible-infrared (RGB-IR) person re-identification (Re-ID) is achieved by projecting them into a common space, allowing Re-ID in 24-hour surveillance systems. However, with respect to probe-to- gallery, almost all existing RGB-IR based methods focus on image-to-image matching, while video-to-video matching which contains much richer spatial- and temporal-information remains under-explored. In this paper, we primarily study video-based per-son...

Image denoising methods using deep neural networks have achieved a great progress in the image restoration. However, recovered images restored by these usually suffer from severe over-smoothness, artifacts, and detail loss. To improve quality of images, we first propose Supplemental Priors (SP) method to adaptively predict depth-directed sample-directed prior information for reconstruction (decoder) networks. Furthermore, over-parameterized too precise supplemental may cause an over-fitting,...

High-resolution automated fingerprint recognition systems (AFRSs) offer higher security because they are able to make use of level-3 features, such as pores, that not available in lower resolution ( <; 500-dpi) images. One the main parameters affecting quality a digital image and issues cost, interoperability, performance an AFRS is choice resolution. In this paper, we identify optimal for using two most representative features: minutiae pores. We first designed multiresolution acquisition...