Weak Segments Research Articles

Weak Edge Target Segmentation Network Based on Dual Attention Mechanism

Segmentation of weak edge targets such as glass and plastic poses a challenge in the field of target segmentation. The detection process is susceptible to background interference and various external factors due to the transparent nature of these materials. To address this issue, this paper introduces a segmentation network for weak edge target objects (WETS-Net). To effectively extract edge information of such objects and eliminate redundant information during feature extraction, a dual-attention mechanism is employed, including the Edge Attention Extraction Module (EAEM) and the Multi-Scale Information Fusion Module (MIFM). Specifically, the EAEM combines improved edge feature extraction kernels to selectively enhance the importance of edge features, aiding in more precise target region extraction. The MIFM utilizes spatial attention mechanisms to fuse multi-scale features, reducing background and external interference. These innovations enhance the performance of WETS-Net, offering a new direction for weak edge target segmentation research. Finally, through ablation experiments, the effectiveness of each module is effectively validated. Moreover, the proposed algorithm achieves an average detection accuracy of 95.83% and 96.13% on the dataset and a self-made dataset, respectively, outperforming similar U-Net-improved networks.

DIAS: A dataset and benchmark for intracranial artery segmentation in DSA sequences

The automated segmentation of Intracranial Arteries (IA) in Digital Subtraction Angiography (DSA) plays a crucial role in the quantification of vascular morphology, significantly contributing to computer-assisted stroke research and clinical practice. Current research primarily focuses on the segmentation of single-frame DSA using proprietary datasets. However, these methods face challenges due to the inherent limitation of single-frame DSA, which only partially displays vascular contrast, thereby hindering accurate vascular structure representation. In this work, we introduce DIAS, a dataset specifically developed for IA segmentation in DSA sequences. We establish a comprehensive benchmark for evaluating DIAS, covering full, weak, and semi-supervised segmentation methods. Specifically, we propose the vessel sequence segmentation network, in which the sequence feature extraction module effectively captures spatiotemporal representations of intravascular contrast, achieving intracranial artery segmentation in 2D+Time DSA sequences. For weakly-supervised IA segmentation, we propose a novel scribble learning-based image segmentation framework, which, under the guidance of scribble labels, employs cross pseudo-supervision and consistency regularization to improve the performance of the segmentation network. Furthermore, we introduce the random patch-based self-training framework, aimed at alleviating the performance constraints encountered in IA segmentation due to the limited availability of annotated DSA data. Our extensive experiments on the DIAS dataset demonstrate the effectiveness of these methods as potential baselines for future research and clinical applications. The dataset and code are publicly available at https://doi.org/10.5281/zenodo.11401368 and https://github.com/lseventeen/DIAS.

Bifunctionalized polyamides based on quinoxaline structure for durable electrochromic display and memory storage

With the aim to obtain multifunctional durable electrochromic (EC) devices, we propose an effective strategy by introducing weak electron acceptor segment with suitable bandgap and oxidation potential. A series of quinoxaline-based units polyamides (PAs) (denoted as T2BPA-CA, T2BPA-OA, T2BPA-TP, and T2BPA-SA) were prepared from a novel donor–acceptor-donor (D-A-D) type diamine with weak electron acceptor quinoxaline and connected with two methoxyl-substituted donor triphenylamine, 10,13-bis[4-aminophenyl (4-methoxyphenyl) amino]-dibenzo[a,c]phenazine, and four dicarboxylic acids, which are high soluble in various organic solvents and exhibit excellent thermostability. Furthermore, the T2BPA-SA thin film exhibits excellent EC performance with outstanding optical contrast (50 %), high coloration efficiency (230 cm2 C−1) and impressive cycle stability (optical contrast remains 80 % over 1400 cycles) in the NIR range. Meanwhile, all of these PAs are served as memristors and exhibit typical nonvolatile ternary memory, realizing an accessible threshold voltage of −0.65 V, a retention time of more than104 s and high current switch ratio (103.52:102.43:1). Overall, this work implies the great potential of these PAs in EC displays and memory storage devices.

Weakly-Supervised Detection of Bone Lesions in CT.

The skeletal region is one of the common sites of metastatic spread of cancer in the breast and prostate. CT is routinely used to measure the size of lesions in the bones. However, they can be difficult to spot due to the wide variations in their sizes, shapes, and appearances. Precise localization of such lesions would enable reliable tracking of interval changes (growth, shrinkage, or unchanged status). To that end, an automated technique to detect bone lesions is highly desirable. In this pilot work, we developed a pipeline to detect bone lesions (lytic, blastic, and mixed) in CT volumes via a proxy segmentation task. First, we used the bone lesions that were prospectively marked by radiologists in a few 2D slices of CT volumes and converted them into weak 3D segmentation masks. Then, we trained a 3D full-resolution nnUNet model using these weak 3D annotations to segment the lesions and thereby detected them. Our automated method detected bone lesions in CT with a precision of 96.7% and recall of 47.3% despite the use of incomplete and partial training data. To the best of our knowledge, we are the first to attempt the direct detection of bone lesions in CT via a proxy segmentation task.

A Soft Prison Called Compulsory School Time

This article attempts to reflect on school time as a paradoxical moment of realization of a right (that of education and upbringing) and de facto negation of other intangible and yet essential rights (the right to play, to the construction of an authentic identity, to self-fulfilment, to freedom). Starting from a provocative, hyperbolic question that could be dismissed as utopian and academically irrelevant: «Can compulsory school time be considered a potential soft crime?», this paper tries to delve into technical and functional aspects of school organization to analyse the potential and shortcomings of en-situation school time, trying to account for the complex dynamic interferences that link functional the normality of the flattening-demotivation of the learners binomial, homologation, social mobility and ordering control. In particular, while citing virtuous and vicious examples of school time taken from realities distant in geographical space or the result of a historical moment other than the current one, this paper focuses on Italy as a case study, with particular reference to the school segment of Kindergarten and some hints relating to primary and lower secondary school - the latter being configured as a weak segment par excellence. Finally, the examination of some legislative measures (in particular Law 53/2003 and Legislative Decree No. 262 of 29 December 2007) will highlight a key problem of school time, namely the split between two moments (infant, primary and middle school on the one hand, upper secondary schools on the other) which is also a split between two modalities, which translates into an opposition between school understood as a social service in the first phase and academia or early professionalization in the second one. A split perpetrated in violation of the healthy principles of didactic continuity and which finds questionable counterweights in the exceptionality of the competition-excellence binomial and in the normality of the flattening-demotivation of the learner’s binomial

An optimized denoised bias correction model with local pre-fitting function for weak boundary image segmentation

The active contour model (ACM) plays a paramount part in grasping visual properties of images and exacting targets of interest. It is overwhelming hardship for traditional ACMs to segment images with noise, intensity inhomogeneity or low contrast and consider computation speed for practical applicability. Therefore, an optimized denoised bias correction (ODBC) model incorporating the pre-piecewise fitting function and the variational denoised term into the energy function is proposed for images with low contrast and intensity inhomogeneity. An optimized gradient descent equation and an upgraded regularization term are produced to enhance the robustness and sensitivity of this model. Experiments are conducted to validate that ODBC model possesses the superiority of reliability and speed in segmenting images with inhomogeneous intensity and strengthens the robustness to initial contours. The results manifest that mIOU of ODBC model exceeds 0.9, superior to prime ACMs and deep learning schemes, in segmenting images associated and unrelated with the dataset.

High resolution magnetic field imaging probe with two rows of TMR sensors for small defects inspection

It is still a challenging problem to detect small defects for eddy current array probes, which requires the probes to possess excellent sensitivity, as well as high spatial resolution. This paper presents a novel high-resolution magnetic field imaging probe with two rows of tunneling magnetoresistance (TMR) array sensors. The bare die sensors are integrated on a printed circuit board by golden wire bonding technology. The two rows of sensors are placed staggered with each other. The data of the two arrays are merged into a matrix, in which way the image pixel pitch is increased to 0.25 mm. The probe employs a differential scheme to suppress the noise, so as to detect the weak signal of small defects. To highlight the weak defect indications, feature extraction and segmentation algorithms are developed. The experimental results confirm that the proposed method can inspect a small defect with dimensions 1 mm (length) × 0.1 mm (width) × 0.1 mm (depth) on a stainless-steel sample.

Impact of Priority Sector Lending in Agriculture: A Study on Select Districts in West Bengal

In India, until the late sixties, commercial banks mainly provided loans to big trade houses for commerce, trade, and business. The stream of loans toward priority sectors, primarily agriculture, weaker segments, and small businesses, is scarce. However, India cannot achieve balanced, equitable, and sustainable growth and development without developing the agriculture sector, given that over 50% of the population is involved. After the nationalization of commercial banks, the broad categories under priority sectors were agriculture, small-scale industry, and exports. The paper relies on primary data about priority sector lending in the agricultural industry of selected blocks in West Bengal to assess the effectiveness of PSL in this sector. The paper employs suitable statistical tools to measure the efficacy. Based on the analysis, results, and findings, the paper offers suggestions to improve the situation of small and marginal farmers and landless farmers, the central portion (around 85% of the agriculture sector) in West Bengal.

Filler effects inspired high performance polyurethane elastomer design: segment arrangement control.

Elastomers with high strength and toughness are in great demand. Previous research on elastomers focused mainly on the design of new chemical structures, but their complicated synthesis process and expensive monomers have restricted the practical application of these materials. Inspired by general filler effects, a strategy is proposed to remarkably enhance the mechanical properties of thermoplastic polyurethane (TPU) elastomers by designing the arrangement of hard/soft segments using traditional chemical compositions. By utilizing the synergetic effect of weak hard segments, normal TPU elastomers are upgraded into advanced elastomers. Combining experiments and simulations, it is demonstrated that a suitable sequence length can achieve considerably enhanced strength and toughness by maximizing the relative surface area of hard domains. Mixing the obtained elastomer with an ionic liquid can result in a durable ionogel sensor with balanced mechanical strength and ionic conductivity. This easy-to-implement strategy offers a new dimension for the development of high-performance elastomers.

Improved U-net network asphalt pavement crack detection method.

Road crack detection is one of the important parts of road safety detection. Aiming at the problems such as weak segmentation effect of basic U-Net on pavement crack, insufficient precision of crack contour segmentation, difficult to identify narrow crack and low segmentation accuracy, this paper proposes an improved U-net network pavement crack segmentation method. VGG16 and Up_Conv (Upsampling Convolution) modules are introduced as backbone network and feature enhancement network respectively, and the more abstract features in the image are extracted by using the Block depth separable convolution blocks, and the multi-scale features are captured and enhanced by higher level semantic information to improve the recognition accuracy of narrow cracks in the road surface. The improved network embedded the Ca(Channel Attention) attention mechanism in U-net's jump connection to enhance the crack portion to suppress background noise. At the same time, DG_Conv(Depthwise GSConv Convolution) module and UnetUp(Unet Upsampling) module are added in the decoding part to extract richer features through more convolutional layers in the network, so that the model pays more attention to the detailed part of the crack, so the segmentation accuracy can be improved. In order to verify the model's ability to detect cracks in complex backgrounds, experiments were carried out on CFD and Deepcrack datasets. The experimental results show that compared with the traditional U-net network F1-score and mIoU have increased by 13.6% and 9.9% respectively. Superior to advanced models such as U-net, Segnet and Linknet in accuracy and generalization ability, the improved model provides a new method for asphalt pavement crack detection. The model is more conducive to practical application and ground deployment, and can be applied in road maintenance projects.

Finding Nano-Ötzi: Cryo-Electron Tomography Visualization Guided by Learned Segmentation.

Cryo-electron tomography (cryo-ET) is a new 3D imaging technique with unprecedented potential for resolving submicron structural details. Existing volume visualization methods, however, are not able to reveal details of interest due to low signal-to-noise ratio. In order to design more powerful transfer functions, we propose leveraging soft segmentation as an explicit component of visualization for noisy volumes. Our technical realization is based on semi-supervised learning, where we combine the advantages of two segmentation algorithms. First, the weak segmentation algorithm provides good results for propagating sparse user-provided labels to other voxels in the same volume and is used to generate dense pseudo-labels. Second, the powerful deep-learning-based segmentation algorithm learns from these pseudo-labels to generalize the segmentation to other unseen volumes, a task that the weak segmentation algorithm fails at completely. The proposed volume visualization uses deep-learning-based segmentation as a component for segmentation-aware transfer function design. Appropriate ramp parameters can be suggested automatically through frequency distribution analysis. Furthermore, our visualization uses gradient-free ambient occlusion shading to further suppress the visual presence of noise, and to give structural detail the desired prominence. The cryo-ET data studied in our technical experiments are based on the highest-quality tilted series of intact SARS-CoV-2 virions. Our technique shows the high impact in target sciences for visual data analysis of very noisy volumes that cannot be visualized with existing techniques.

A Novel Gradient-Weighted Voting Approach for Classical and Fuzzy Circular Hough Transforms and Their Application in Medical Image Analysis—Case Study: Colonoscopy

Classical circular Hough transform was proven to be effective for some types of colorectal polyps. However, the polyps are very rarely perfectly circular, so some tolerance is needed, that can be ensured by applying fuzzy Hough transform instead of the classical one. In addition, the edge detection method, which is used as a preprocessing step of the Hough transforms, was changed from the generally used Canny method to Prewitt that detects fewer edge points outside of the polyp contours and also a smaller number of points to be transformed based on statistical data from three colonoscopy databases. According to the statistical study we performed, in the colonoscopy images the polyp contours usually belong to gradient domain of neither too large, nor too small gradients, though they can also have stronger or weaker segments. In order to prioritize the gradient domain typical for the polyps, a relative gradient-based thresholding as well as a gradient-weighted voting was introduced in this paper. For evaluating the improvement of the shape deviation tolerance of the classical and fuzzy Hough transforms, the maximum radial displacement and the average radius were used to characterize the roundness of the objects to be detected. The gradient thresholding proved to decrease the calculation time to less than 50% of the full Hough transforms, and the number of the resulting circles outside the polyp’s environment also decreased, especially for low resolution images.

Enabling Multi-Part Plant Segmentation with Instance-Level Augmentation Using Weak Annotations

Plant segmentation is a challenging computer vision task due to plant images complexity. For many practical problems, we have to solve even more difficult tasks. We need to distinguish plant parts rather than the whole plant. The major complication of multi-part segmentation is the absence of well-annotated datasets. It is very time-consuming and expensive to annotate datasets manually on the object parts level. In this article, we propose to use weakly supervised learning for pseudo-annotation. The goal is to train a plant part segmentation model using only bounding boxes instead of fine-grained masks. We review the existing weakly supervised learning approaches and propose an efficient pipeline for agricultural domains. It is designed to resolve tight object overlappings. Our pipeline beats the baseline solution by 23% for the plant part case and by 40% for the whole plant case. Furthermore, we apply instance-level augmentation to boost model performance. The idea of this approach is to obtain a weak segmentation mask and use it for cropping objects from original images and pasting them to new backgrounds during model training. This method provides us a 55% increase in mAP compared with the baseline on object part and a 72% increase on the whole plant segmentation tasks.

Protective Discrimination in Favour of Backward Classes under Indian Constitution

The oppression of the weaker segment in the society is a well known fact to all people. The subjection of the weaker segment of the society to one or the other kind of discrimination and suppression at the hands of the advanced and the society’s dominant sections are as old as Adam. They had been suffering a lot in each and every day throughout their lives. They had been made subject to discrimination in one way or the other at stages in their life. This evil had to be nipped in the bud, however now as aftermath, it has ended up a massive tree with its roots deeply buried in the Indian society. The higher crust rules and regulations of democracy are everywhere. The framers and makers of our constitution were thoroughly conscious about the distressed condition of these weaker sections. They tried to put their best foot forward to lift the weaker segment by offering a few special provisions within the constitution like Equality under Article 14 and Article 15 prohibiting discrimination and many more. In a number of the Articles, the framers of our constitution have provided additional privileges in favor of the weak and backward segment of the society. In other words, it means that the makers of the constitution have provided protective discrimination within the constitution.

Parameters Modeling and Probing of Highway Structural Deterioration: Case study of F-209 Segment of Ondo - Ore Pavement, Southwestern Nigeria

The construction of highway route is influenced by geology, geotechnical, topography, and geomorphology of the terrain. However, for pavement that failed structurally, a critical investigation is required to ascertain the cause(s) of its failure, so that such study can assist during the rehabilitation/re-construction phase. Based on this, the incessant failed Ondo – Ore was studied using integrated methods. Electrical resistivity involving four vertical electrical sounding, ten dynamic cone penetration test, three coring by trial pits at 1.0 to 3.0 m offset from the edge of the highway at different chain-age, and laboratory geotechnical-geochemical analysis adopting standard procedures. The VES characterized the geological sequence within the highway alignment to be topsoil, subsoil, weathered layer, and basement rock. The topsoil/subsoil and the weathered layer on which the highway is founded is sandy clay and clay with resistivity less than 200 ohm-m. The depths to basement ranged from 33.5 to 45.1 m. The trial pits recordings are consistent with the results of the VES, geotechnical, and geochemical analysis, distinctly distinguishing the upper 1 m into clayey soil (sandy clay, clayey hardpan) laterite, and silt-clay-sand mixture. The engineering competence of the topsoil/subsoil on which the road is founded is poor, although is inactive SC-SM, A-7-5/A-7-6 lateritic soil type (silica-sesquioxide ratio of 1.67). The clay mineralogy is within the illite – montmorillonite group. The SNG, SN, and SNP contributions of the soil as subgrade, subbase, and base material are very low (<1.0). The regression models of all parameters correlated positively, although weak for RD and DCPI, in-situ CBR and MR, RD and in-situ CBR; while strong for soaked CBR and in-situ CBR, ER and MR. Consequently, based on the GI and CBR values, the expected average thickness of the highway should range between 191 mm (good segment) to 445 mm (for weak segment) (avg. 312 mm) which is far thicker than 274 mm measured along the highway alignment during reconnaissance survey. Therefore, it can be concluded that the failure of the highway is as a result of low soaked CBR/in-situ CBR values with low strength coefficient; and low design thickness across the highway. In addition, lack of drainage at the shoulders of the highway is also causative factor. Hence, the need for effective design of roads (to specification) and maintenance strategy was therefore advocated.

Customized RBF kernel graph-cut for weak boundary image segmentation

Customized RBF kernel graph-cut for weak boundary image segmentation

A sub-region Unet for weak defects segmentation with global information and mask-aware loss

In recent years, diverse detection methods have been proposed to achieve highly accurate image segmentation. However, low contrast and background noise interference still pose challenges to defect detection. Existing deep learning-based methods for defect detection mostly learn rich features on the whole image. This can lead to overfitting as the parameters get bigger, and it also loses the spatial connection. In order to solve this problem, we suggest a network based on spatial sub-region feature extraction and mask-aware loss that can detect metal surface defects. In this network framework, we obtain the contextual information of each region through two streams: one branch encodes the spatial sub-region information and reduces the effect of geometric and illumination deformation under u-shape architecture. A skip-connected structure in another stream is designed to learn global information and compensate for boundary features. Next, we propose a mask-aware loss to reduce background noise interference further. Finally, we validate the effectiveness of our method on two challenging datasets, reaching 88.45% and 79.16% on the NEU-Seg and USB-Seg datasets, respectively. Also, our model is only 5.48 MB, and can achieve 172 fps for processing images with 200 × 200 size defects.

A Weakly Supervised-Guided Soft Attention Network for Classification of Intracranial Hemorrhage

Intracranial hemorrhage (ICH) classification from computed tomography (CT) scans is important for accurate diagnosis of stroke in emergency centers. However, it is challenged by low image contrast and the complex appearance of the lesion. Therefore, the diagnosis results may vary from doctor to doctor under different situations. Existing auxiliary diagnosis algorithms treat the features of each location as equally important, and they do not impose any constraints on the learned features. The learned features used for classification in the model contain noise or features that have no diagnostic significance, which greatly limits the performance and reliability of systems. To facilitate the ICH treatment, we propose to use the auxiliary weak-segmentation-label supervision in multiscale CNN to classify ICH lesions on brain CT images, where the ICH lesions are segmented and used to guide the attention of the classification network. The weak segmentation labels we use are generated by an unsupervised algorithm and do not require additional annotations. We conduct experiments on the RSNA ICH detection data set, and the results demonstrate that our method can achieve an average ACC of 98.1% and an average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> score of 74.6% on the test set of nearly 80 000 CT images. The proposed weakly supervised-guided attention mechanism can accurately activate those neurons related to diagnosis, while inhibiting the activation of irrelevant areas and noise, thus achieving good performance.

Multiple Temporal Pooling Mechanisms for Weakly Supervised Temporal Action Localization

Recent action localization works learn in a weakly supervised manner to avoid the expensive cost of human labeling. Those works are mostly based on the Multiple Instance Learning framework, where temporal pooling is an indispensable part that usually relies on the guidance of snippet-level Class Activation Sequences (CAS) . However, we observe that previous works only leverage a simple convolutional neural network for the generation of CAS, which ignores the weak discriminative foreground action segments and the background ones, and meanwhile, the relationship between different actions has not been considered. To solve this problem, we propose multiple temporal pooling mechanisms (MTP) for a more sufficient information utilization. Specifically, with the design of the Foreground Variance Branch, Dual Foreground Attention Branch and Hybrid Attention Fine-tuning Branch, MTP can leverage more effective information from different aspects and generate different CASs to guide the learning of temporal pooling. Moreover, different loss functions are designed for a better optimization of individual branches, aiming to effectively distinguish the action from the background. Our method shows excellent results on the THUMOS14 and ActivityNet1.2 datasets.

Attention-effective multiple instance learning on weakly stem cell colony segmentation

The detection of induced pluripotent stem cell (iPSC) colonies often needs the precise extraction of the colony features. However, existing computerized systems relied on segmentation of contours by preprocessing for classifying the colony conditions were task-extensive. To maximize the efficiency in categorizing colony conditions, we propose a multiple instance learning (MIL) in weakly supervised settings. It is designed in a single model to produce weak segmentation and classification of colonies without using finely labeled samples. As a single model, we employ a U-net-like convolution neural network (CNN) to train on binary image-level labels for MIL colonies classification. Furthermore, to specify the object of interest we used a simple post-processing method. The proposed approach is compared over conventional methods using five-fold cross-validation and receiver operating characteristic (ROC) curve. The maximum accuracy of the MIL-net is 95%, which is 15% higher than the conventional methods. Furthermore, the ability to interpret the location of the iPSC colonies based on the image level label without using a pixel-wise ground truth image is more appealing and cost-effective in colony condition recognition.