Most attention-embedded networks fall short in effectively integrating spatial / channel-wise information across diverse scales, leading to suboptimal performance for coronary vessels segmentation in X-ray digital subtraction angiography images. To address this limitation, a multi-scale across attention incorporated network (named MS2A-Net) is introduced. MS2A-Net accepts original and enhanced images as inputs, leveraging complementary information provided by the different contrasts within the images. Furthermore, MS2A is designed to integrate features across multiple levels, scales and sources, for effectively extracting deep semantic information. After the incorporation of features with adaptive weightings, segmentation accuracy is refined. Qualitative and quantitative experiment results prove that MS2A-Net not only outperforms state-of-the-art tactics but also shows superior performance, e.g., higher intersection over union, Dice coefficient, broader areas under receiver operating characteristic curves.

As the depth of the detection network increases, features of small targets become less pronounced, and the model may inadvertently favor background clutter, thereby reducing the efficiency of target detection. Hence, a bidimensional multilevel attention-embedded network, named BiMA, is designed to simultaneously tackle original images rich in detail and filtered images that remove superfluous non-target elements, serving as its dual-input framework. Subsequently, a multi-scale fusion is raised to harness multi-scale properties of these images, offering a thorough grasp of both high-level and low-level features. In addition, a multi-level attention is integrated to further improve the detection capability. Extensive qualitative and quantitative experiments prove that BiMA not only surpasses contemporary approaches in robustly detecting small targets across a range of challenging scenarios but also has superior performance, e.g., higher probabilities of detection, reduced false alarm rates, and broader areas under receiver operating characteristic curves.

The research of single-frame infrared small target detection has made some achievements, but still needs to be improved. Traditional methods rely on the assumption of target background difference, and the effect is poor in multi-scale small target detection. Although the deep learning method is more promising, it is still affected by the complex background noise and the loss of small targets in the deep network. Therefore, we propose DMSnet, a dual multi-scale background stripping nested network. Firstly, the detection is transformed into a low rank sparse matrix optimization problem to strip part of the adverse information by using the non-local self-similarity of the infrared small target image. Then the original image with complete and rich details and the improved image with stripped background and some non-target elements are input into the double branch structure for feature extraction. Feature extraction uses a nested u-net structure, which can retain the representation of small targets at a deeper level. It combines with the spatial and channel attention module for feature enhancement. Finally, multi-scale fusion is used to make full use of high-level and low-level image feature information. Experiments on NUDT-SIRST and SIRST datasets show that DMSnet stably surpasses a large number of existing methods in terms of IoU, PD and FA, indicating the effectiveness of DMSnet in detecting small targets in infrared images with complex backgrounds.

Accurate drug dosage in cancer treatments depends on precise liver and tumor size estimation, but tumors pose challenges for automatic segmentation due to their small, scattered volumes, complex structures, and low contrast against surrounding organs. Conventional U-Net models struggle to capture these microstructural details. To address this, we propose LKC-Net, a liver and tumor segmentation method leveraging large kernel parallel dilated convolutions. The encoder utilizes depth-wise separable large kernels combined with parallel dilated convolutions and squeeze-and-excitation (SE) modules. This architecture enhances the receptive field and improves the capture of small-scale patterns. SE modules filter redundant information, focusing the model on key regions. Experiments on ATLAS, LiTS, and private datasets demonstrate LKC-Net’s superior performance in liver and tumor segmentation tasks.

When infrared small-dim target images under strong background clutters are employed to train a deep learning-based detection network, the model becomes biased towards the clutters, negatively impacting detection performance. While background estimation is able to address this issue, most convolutional neural network-based ways require manual foreground mask extraction during learning phases. In unsupervised tactics, it is often assumed that the background in frames is captured by a stationary camera. Howbeit, lots of small target sequences have dynamic backgrounds due to motion in the imaging platform, challenging this hypothesis. There has limited focus on unsupervised background estimation for small target images with sensor motion. To address this gap, a learning-based model, named BEmST, is raised. BEmST combines a variational autoencoder with stable principal component pursuit optimization for unsupervised deep background modelling. Target detection is then performed using U-net++ on differences between the modelled background and input images. This innovative tactic integrates unsupervised probabilistic background estimation with supervised dense classification for bettered small target detection. Extensive qualitative / quantitative experiments on public datasets validate that BEmST not only outperforms state-of-the-art tactics in availably and robustly detecting small-dim target images across various challenging scenarios, but also achieves superior detection performance, such as higher probabilities of detection, lower false alarm rates, and larger areas under ROC curves. The results pave a way for the future utilization of small-dim target image in a more efficient manner.

It is challenging to precisely segment the liver from surrounding organs in medical images because of the poor contrast between them. A method of semantic segmentation of 3D liver images based on multi-path features attention mechanism is proposed to address this issue. It integrates three-dimensional spatial information and feature information from several paths in the model to automatically segment the liver area. The model in this paper uses the LiTS dataset for training, testing, and ablation experiments, and compares the results with previous models. The experimental results demonstrate that the model in this paper has reached 0.965 in the DICE similarity coefficient, and has also improved in evaluation indicators such as volume overlap error (VOE) and root mean square symmetric surface distance (RMSD). It also has better segmentation performance when tested on the CHAOS dataset. Cross-validation was carried out on the clinical MRI dataset of a hospital, and the DICE similarity coefficient reached 0.971. The results show that the model has good performance on the multi-modal datasets of CT and MRI.

Oriented object detectors provide many scientific solutions for object detection tasks in remote sensing scenes. Among them, anchor-free oriented object detectors have attracted much attention because of their flexibility and conciseness in recent years. However, the positive sampling strategy adopted by most modern anchor-free oriented detectors is inadequate to reflect remote sensing objects characterized by arbitrary orientation, densely packed and extensive scale variation. Besides, there is still a lack of a general representation to describe oriented objects shaped with arbitrary convex quadrilaterals. In this paper, we present a one-stage anchor-free oriented object detector, called M2VDet, which focuses on designing the sampling strategy and oriented bounding box description. First, an adaptive sampling strategy is proposed to generate positive sample spaces, which take full account of objects’ characteristics like scale, shape, and orientation. The resulting candidate regions provide a superior space constraint to construct the heatmap, then 2D oriented Gaussian distribution is employed to generate prior labels for positive candidates. Second, a general oriented bounding box representation is designed by adopting the midpoints-to-vertices strategy, which provides a unified approach to describe the arbitrary convex quadrilaterals without brutal approximation. Extensive experimental results on four public datasets (e.g., DOTA, HRSC2016, UCAS-AOD, and SODA-A) demonstrate that M2VDet is with a simple pipeline, but it is able to achieve more robust and available performance when compared with state-of-the-art baseline detectors, especially on the detection of densely packed objects with arbitrary orientation as well as small objects.

Traditional matrix-based approaches in the field of finite state machines construct state transition matrices, and then use the powers of the state transition matrices to represent corresponding dynamic transition processes, which are cornerstones of system analysis. In this study, we propose a static matrix-based approach that revisits a finite state machine from its structure rather than its dynamic transition process, thus avoiding the “explosion of complexity” problem inherent in the existing approaches. Based on the static approach, we reexamine the issues of closed-loop detection and controllability for deterministic finite state machines. In addition, we propose controllable equivalent form and minimal controllable equivalent form concepts and give corresponding algorithms.

Motivated by the inconvenience or even inability to explain the mathematics of the state space optimization of finite state machines (FSMs) in most existing results, we consider the problem by viewing FSMs as logical dynamic systems. Borrowing ideas from the concept of equilibrium points of dynamic systems in control theory, the concepts of t -equivalent states and t -source equivalent states are introduced. Based on the state transition dynamic equations of FSMs proposed in recent years, several mathematical formulations of t -equivalent states and t -source equivalent states are proposed. These can be analogized to the necessary and sufficient conditions of equilibrium points of dynamic systems in control theory and thus give a mathematical explanation of the optimization problem. Using these mathematical formulations, two methods are designed to find all the t -equivalent states and t -source equivalent states of FSMs. Further, two ways of reducing the state space of FSMs are found. These can be implemented without computers but with only pen and paper in a mathematical manner. In addition, an open question is raised which can further improve these methods into unattended ones. Finally, the correctness and effectiveness of the proposed methods are verified by a practical language model.

In a low signal-to-clutter ratio (SCR) small-infrared-target image with chaotic cloudy-/sea-sky background, the target has very similar thermal intensities to the background (e.g., edges of clouds). In such case, how to accurately detect small targets is crucial in infrared search and tracking applications. Conventional methods based on the local difference/mutation potentially result in high miss and/or false alarm rates. Here, we propose an effective method for detecting small infrared targets embedded in complex backgrounds through a multiscale fuzzy metric that measures the certainty of targets in images. Accordingly, the detection task is formulated as a fuzzy measure issue. The presented metric is able to eliminate substantial background clutters and noise. Especially, it significantly improves SCR values of the image. Subsequently, a simple and adaptive threshold is used to segment target. Extensive clipped and real data experiments demonstrate that the proposed algorithm not only works more robustly for different target sizes, SCR values, target and/or background types, but also has better performance regarding detection accuracy, when compared with traditional baseline methods. Moreover, the mathematical proofs are provided for understanding the proposed detection method.

Xe) dynamic MRI could visualize the lung ventilation process, which provides characteristics regarding lung physiology and pathophysiology. Compressed sensing (CS) is generally used to increase the temporal resolution of such dynamic MRI. Nevertheless, the acceleration factor of CS is constant, which results in difficulties in precisely observing and/or measuring dynamic ventilation process due to bifurcating network structure of the lung. Here, an adaptive strategy is proposed to highly undersample pulmonary HP dynamic k-space data, according to the characteristics of both lung structure and gas motion. After that, a valid reconstruction algorithm is developed to reconstruct dynamic MR images, considering the low-rank, global sparsity, gas-inflow effects, and joint sparsity. Both the simulation and the in vivo results verify that the proposed approach outperforms the state-of-the-art methods both in qualitative and quantitative comparisons. In particular, the proposed method acquires 33 frames within 6.67 s (more than double the temporal resolution of the recently proposed strategy), and achieves high-image quality [the improvements are 29.63%, 3.19%, 2.08%, and 13.03% regarding the mean absolute error (MAE), structural similarity index (SSIM), quality index based on local variance (QILV), and contrast-to-noise ratio (CNR) comparisons]. This provides accurate structural and functional information for early detection of obstructive lung diseases.

Enhancement of images with weak edges faces great challenges in imaging applications. In this study, the authors propose a novel image enhancement approach based on intuitionistic fuzzy sets. The proposed method first divides an image into sub‐object and sub‐background areas, and then successively implements new fuzzification, hyperbolisation, and defuzzification operations on each area. In this way, an enhanced image is obtained, where the visual quality of region of interest (ROI) is significantly improved. Several types of images are utilised to validate the proposed method with respect to the enhancement performance. Experimental results demonstrate that the proposed algorithm not only works more stably for different types of images, but also has better enhancement performance, in comparison to conventional methods. This is a great merit of such design for discerning specific ROIs.

Against an intricate infrared cloudy-sky background, jamming objects such as the edges of clouds in the scene have a similar thermal intensity measure with respect to the background as small targets. This may cause high false alarm rates and low probabilities of detection according to conventional small target detection methods. In this paper, we propose a weighted local difference measure (WLDM)-based scheme for the detection of small targets against various complex cloudy-sky backgrounds. Initially, a WLDM map is achieved to simultaneously enhance targets and suppress background clutters and noise. In this way, the true targets can be easily separated from jamming objects. After that, a simple adaptive threshold is used to segment the targets. More than 460 infrared small target images against diverse intricate cloudy-sky backgrounds were utilized to validate the detection capability of the WLDM-based method. Experimental results demonstrate that the proposed algorithm not only works more robustly for different cloudy-sky backgrounds, target movements, and signal-to-clutter ratio (SCR) values but also has a better performance with regard to the detection accuracy, in comparison to traditional baseline methods. In particular, the proposed method is able to significantly improve SCR values of the images.

Background suppression is vitally important for the small target detection, which aims to enhance targets and improve the signal‐to‐noise ratio of small target images. Consequently, the study proposes a background suppression approach based on the fast local reverse entropy operator, which is designed according to the fact that the appearance of a small target could result in the great change of the value of local reverse entropy in the local region. The operator is adopted to suppress complex backgrounds of small target images in order to enhance small targets, and then bring about high probabilities of detection and low probabilities of false alarm in the small target detection. Both quantitative and qualitative analyses contribute to confirm the validity and efficiency of the proposed approach.