Image Splitting Research Articles

Experimental investigation on mechanical properties and characterization of steel fibre concrete with Bacillus subtilis

Building cracks spoil the aesthetic view of the structure along with degrading the strength of the structure. It leads to the failure of the structure as a whole. Propagation of cracks in the concrete surface increases the chance of permeability due to moisture content in the atmosphere which might corrode the internal reinforcements. To overcome this situation, a solution in the form of fibre-reinforced self-healing concrete was suggested in this research work. Steel fibre was added by varying 0%,1%,2%,3%,4% and 5% interms of volume of concrete to prolong fatigue life and decrease the crack width under fatigue loading. Bacteria (Bacillus subtilis) are used to heal the cracks by producing calcium carbonate (CaCO3) as a result. Bacterial concrete with fibre (BCF) was cast by M30 mix as per Indian Standard Code. Studied concrete's mechanical and microstructural properties like compression strength test, flexural strength test, split tensile strength test, SEM images, EDS, XRD and FTIR methods. From microstructural studies, it is clear that there is sufficient self-healing material in broken concrete, hence the efficiency of BCF's self-healing property is commendable and the ability of the generated BCF concrete to regain strength.

Unsupervised Adversarial Domain Adaptation for Agricultural Land Extraction of Remote Sensing Images

Agricultural land extraction is an essential technical means to promote sustainable agricultural development and modernization research. Existing supervised algorithms rely on many finely annotated remote-sensing images, which is both time-consuming and expensive. One way to reduce the annotation cost approach is to migrate models trained on existing annotated data (source domain) to unannotated data (target domain). However, model generalization capability is often unsatisfactory due to the limit of the domain gap. In this work, we use an unsupervised adversarial domain adaptation method to train a neural network to close the gap between the source and target domains for unsupervised agricultural land extraction. The overall approach consists of two phases: inter-domain and intra-domain adaptation. In the inter-domain adaptation, we use a generative adversarial network (GAN) to reduce the inter-domain gap between the source domain (labeled dataset) and the target domain (unlabeled dataset). The transformer with robust long-range dependency modeling acts as the backbone of the generator. In addition, the multi-scale feature fusion (MSFF) module is designed in the generator to accommodate remote sensing datasets with different spatial resolutions. Further, we use an entropy-based approach to divide the target domain. The target domain is divided into two subdomains, easy split images and hard split images. By training against each other between the two subdomains, we reduce the intra-domain gap. Experiments results on the “DeepGlobe → LoveDA”, “GID → LoveDA” and “DeepGlobe → GID” unsupervised agricultural land extraction tasks demonstrate the effectiveness of our method and its superiority to other unsupervised domain adaptation techniques.

Brain-like retinex: A biologically plausible retinex algorithm for low light image enhancement

Retinex theory was first proposed by Land and McCann [1], where retinex is a portmanteau derived from the words of retina and cortex, implying that both the retina and cerebral cortex may participate in the perception of lightness and color. However, there are no recent reports on how the retina and visual cortex perform retinex decomposition. In this paper, we propose a biologically plausible solution to retinex decomposition. We develop an algorithm motivated by the primate’s retinal circuit to detect textural gradients, design an algorithm originating from the visual cortex to extract image contours, and thus split image edges into image contours and textural gradients. Then, we establish a variational model for retinex decomposition by using image contours and textural gradients to encode discontinuities in illumination and variations in reflectance, respectively. We also apply the proposed retinex model to low light image enhancement, high dynamic resolution image toning, and color constancy. Experiments show consistent superiority of the proposed algorithm. The code is available at Github.

Study of Molecular Cytogenetic Abnormalities in Lymphoma and Their Clinical Relevance

Objective: To investigate the issue of molecular cytogenetic abnormalities in lymphoma and to study their clinical relevance. Methods: First, medium-term chromosomes of healthy men and women were prepared. Normal human peripheral blood lymphocytes were stimulated with PHA, treated with MTX synchronization, blocked with colchicine, cultured for 3 days, routinely treated with hypotonicity, cells were obtained, and steam-dried method was used to drop slices for backup. Next, comparative genomic hybridization. Preparation of probes; processing of normal human mid-term split images; hybridization; washing of slices after hybridization; staining; fluorescence microscopy observation, uptake and analysis of fluorescence signals. Results: Lymphoma CGH test results; Relationship between CGH abnormalities and lymphoma histological subtypes:Hodgkin's lymphoma CGH abnormalities, diffuse large B-cell lymphoma (DLBCL) CGH abnormalities, follicular lymphoma (FL) CGH abnormalities, and follicular lymphoma (FL) CGH abnormalities; Correlation between CGH test results and lymphoma clinical features: CGH test results and lymphoma patients, the type of CGH abnormality correlated with the clinical features of lymphoma, simple CGH abnormalities and complex CGH abnormalities correlated with the clinical features of lymphoma patients, and CGH abnormalities at specific chromosomal sites correlated with the clinical features of lymphoma. Conclusion:Conventional model analysis examines both chromosomal translocations and gene deletions or spreads, fails to observe the full picture of the lymphoma genome, and fails to detect new chromosomal abnormalities.

Deep and Lightweight Neural Network for Histopathological Image Classification

Breast cancer is a fatal disease affecting women, and early detection and proper treatment are crucial. Classifying medical images correctly is the first and most important step in the cancer diagnosis stage. Deep learning-based classification methods in various domains demonstrate advances in accuracy. However, as deep learning improves, the layers of neural networks get deeper, raising challenges, such as overfitting and gradient vanishing. For instance, a medical image is simpler than an ordinary one, making it vulnerable to overfitting issues. We present breast histopathological classification methods with two deep neural networks, Xception and LightXception with aid of voting schemes over split images. Most deep neural networks classify thousands classes of images, but the breast histopathological image classes are far fewer than those of other image classification tasks. Because the BreakHis dataset is relatively simpler than typical image datasets, such as ImageNet, applying the conventional highly deep neural networks may suffer from the aforementioned overfitting or gradient vanishing problems. Additionally, highly deep neural networks require more resources, leading to high computational costs. Consequently, we propose a new network; LightXception by cutting off layers at the bottom of the Xception network and reducing the number of channels of convolution filters. LightXception has only about 35% of parameters compared to those of the original Xception network with minimal expense on performance. Based on images with 100X magnification factor, the performance comparisons for Xception vs. LightXception are 97.42% vs. 97.31% on classification accuracy, 97.42% vs. 97.42% on recall, and 99.26% vs. 98.67% of precision.

NRT-YOLO: Improved YOLOv5 Based on Nested Residual Transformer for Tiny Remote Sensing Object Detection.

To address the problems of tiny objects and high resolution of object detection in remote sensing imagery, the methods with coarse-grained image cropping have been widely studied. However, these methods are always inefficient and complex due to the two-stage architecture and the huge computation for split images. For these reasons, this article employs YOLO and presents an improved architecture, NRT-YOLO. Specifically, the improvements can be summarized as: extra prediction head and related feature fusion layers; novel nested residual Transformer module, C3NRT; nested residual attention module, C3NRA; and multi-scale testing. The C3NRT module presented in this paper could boost accuracy and reduce complexity of the network at the same time. Moreover, the effectiveness of the proposed method is demonstrated by three kinds of experiments. NRT-YOLO achieves 56.9% mAP0.5 with only 38.1 M parameters in the DOTA dataset, exceeding YOLOv5l by 4.5%. Also, the results of different classifications show its excellent ability to detect small sample objects. As for the C3NRT module, the ablation study and comparison experiment verified that it has the largest contribution to accuracy increment (2.7% in mAP0.5) among the improvements. In conclusion, NRT-YOLO has excellent performance in accuracy improvement and parameter reduction, which is suitable for tiny remote sensing object detection.

Reverse split hand syndrome and distinctive spine imaging features in Hirayama disease.

Reverse split hand syndrome and distinctive spine imaging features in Hirayama disease.

Development of a split fluorescent protein-based RNA live-cell imaging system to visualize mRNA distribution in plants

BackgroundRNA live-cell imaging systems have been used to visualize subcellular mRNA distribution in living cells. The RNA-binding protein (RBP)-based RNA imaging system exploits specific RBP and the corresponding RNA recognition sequences to indirectly label mRNAs. Co-expression of fluorescent protein-fused RBP and target mRNA conjugated with corresponding RNA recognition sequences allows for visualizing mRNAs by confocal microscopy. To minimize the background fluorescence in the cytosol, the nuclear localization sequence has been used to sequester the RBP not bound to mRNA in the nucleus. However, strong fluorescence in the nucleus may limit the visualization of nucleus-localized RNA and sometimes may interfere in detecting fluorescence signals in the cytosol, especially in cells with low signal-to-noise ratio.ResultsWe eliminated the background fluorescence in the nucleus by using the split fluorescent protein-based approach. We fused two different RBPs with the N- or C-terminus of split fluorescent proteins (FPs). Co-expression of RBPs with the target mRNA conjugated with the corresponding RNA recognition sequences can bring split FPs together to reconstitute functional FPs for visualizing target mRNAs. We optimized the system with minimal background fluorescence and used the imaging system to visualize mRNAs in living plant cells.ConclusionsWe established a background-free RNA live-cell imaging system that provides a platform to visualize subcellular mRNA distribution in living plant cells.

Spatial and topology feature extraction on batik pattern recognition: a review

. Batik is an Indonesian cultural heritage that has been recognized by UNESCO as an international cultural heritage on October 2, 2009. Patterns of batik produce geometric shapes unique, the number and name of the batik patterns make it difficult to recognize each motif. The objective classification of batik is split image into classes according to the pattern motif motive so easy to recognize in accordance with its feature. Batik can be classified based on the shape of the motive, namely geometric motifs, geometric motifs and motifs non specific. Spatial information is an important aspect of image processing such as computer vision and recognition structure / pattern in the context of modelling and resolution of the uncertainty caused by the ambiguity in the low-level features. Shortcomings inherent in combining two colours and spatial features are not adaptive pattern recognition process of the region across multiple images and histogram matching is not appropriate to capture the colours on the image content. This study discussed a model of spatial features and feature combinations topology with the aim to improve the validation batik image pattern recognition so that the level of the pattern recognition motif batik image could be better. Some of the features that have been used include colour features and spatial features. In addition, this paper discusses the possibility of combining the features in pattern recognition. This paper proposes a combination of features that will be able to improve the validation of image pattern recognition of batik.

CNN BASED METHOD FOR MULTI-TYPE DISEASED ARECANUT IMAGE CLASSIFICATION

Arecanut image classification is a challenging task to the researchers and in this paper a new combined approach of multi-gradient images and deep convolutional neural networks for multi-type arecanut image classification is presented. To enhance the fine details in arecanut images affected by different diseases, namely, rot, split and rot-split, we propose to explore multiple-Sobel masks for convolving with the input image. Although, the images suffer from distortion due to disease infection, this masking operation helps to enhance the fine details. We believe that the fine details provide vital clues for classification of normal, rot, split and rot-split images. To extract such clues, we explore the combination of multi-gradient and AlexNet by feeding enhanced images as input. Implementation results on the four-class dataset indicate that the approach proposed is superior in terms of classification rate, recall, precision and F-measures. The same conclusion can be drawn from the results of comparative study of proposed method with the existing methods.

Description-based person search with multi-grained matching networks

Description-based person search aims to retrieve a person in the image database based on a description about that person. It is a challenging task since the visual image and the textual description belong to different modalities. To fully capture the relevance between person images and textual descriptions, we propose a multi-grained framework with three branches for visual-textual matching. Specifically, in the global-grained branch, we extract global contexts from the entire images and descriptions. In the fine-grained branch, we adopt visual human parsing and linguistic parsing to split images and descriptions into semantic components related to different body parts. We design two attention mechanisms including segmentation-based and linguistics-based attention to align visual and textual semantic components for fine-grained matching. To further exploit the spatial relations between fine-grained semantic components, we construct a body graph in the coarse-grained branch and exploit graph convolutional neural networks to aggregate fine-grained components into coarse-grained representations. The visual and textual representations learned by three branches are complementary to each other which enhance the visual-textual matching performance. Experimental results on the CUHK-PEDES dataset show that our approach performs favorably against state-of-the-art description-based person search methods.

Actin-Related Protein 4 Interacts with PIE1 and Regulates Gene Expression in Arabidopsis.

As essential structural components of ATP-dependent chromatin-remodeling complex, the nucleolus-localized actin-related proteins (ARPs) play critical roles in many biological processes. Among them, ARP4 is identified as an integral subunit of chromatin remodeling complex SWR1, which is conserved in yeast, humans and plants. It was shown that RNAi mediated knock-down of Arabidopsis thaliana ARP4 (AtARP4) could affect plant development, specifically, leading to early flowering. However, so far, little is known about how ARP4 functions in the SWR1 complex in plant. Here, we identified a loss-of-function mutant of AtARP4 with a single nucleotide change from glycine to arginine, which had significantly smaller leaf size. The results from the split luciferase complementation imaging (LCI) and yeast two hybrid (Y2H) assays confirmed its physical interaction with the scaffold and catalytic subunit of SWR1 complex, photoperiod-independent early flowering 1 (PIE1). Furthermore, mutation of AtARP4 caused altered transcription response of hundreds of genes, in which the number of up-regulated differentially expressed genes (DEGs) was much larger than those down-regulated. Although most DEGs in atarp4 are related to plant defense and response to hormones such as salicylic acid, overall, it has less overlapping with other swr1 mutants and the hta9 hta11 double-mutant. In conclusion, our results reveal that AtARP4 is important for plant growth and such an effect is likely attributed to its repression on gene expression, typically at defense-related loci, thus providing some evidence for the coordination of plant growth and defense, while the regulatory patterns and mechanisms are distinctive from other SWR1 complex components.

Dynamic diagnosis of early gastric cancer with microvascular blood flow rate using magnifying endoscopy (with video): A pilot study.

Background and AimMagnifying endoscopy (ME) diagnostic algorithm for early gastric cancer (EGC) relies on qualitative features such as microvascular (MV) architecture and microsurface structure; however, it is a “static” diagnostic algorithm that uses still images. ME can visualize red blood cell flow within subepithelial microvessels in real time. Here, we evaluated the utility of using the MV blood flow rate in combination with ME for the diagnosis of EGC as a retrospective study.MethodsPatients with differentiated‐type EGC (n = 10) or patchy redness (n = 10) underwent ME with blue laser imaging. The mean MV blood flow rates of EGC, patchy redness, and background mucosa were calculated by the mean movement distance of one tagging red blood cell using split images of ME with blue laser imaging videos. We compared the mean MV blood flow rate between EGC, patchy redness, and background mucosa and also calculated the MV blood flow imaging ratio (inside lesion/background mucosa) between EGC and patchy redness.ResultsMean MV blood flow rate was significantly lower in EGC (1481 μm/s; range 1057–1762) than in patchy redness (3859 μm/s; 2435–5899) or background mucosa (4140.6 μm/s; 2820–6247) (P < 0.01). The MV blood flow imaging ratio was significantly lower in EGC (0.39; 0.27–0.62) than in patchy redness (0.90; 0.78–1.1) (P < 0.01).ConclusionsDynamic diagnosis with MV blood flow rate using ME may be useful for the differential diagnosis of EGC and patchy redness. Endoscopic assessment of dynamic processes within the gastric mucosa may facilitate the diagnosis of EGC.

Comparison of the Methods of Image Slicing After Initial Image Processing Using the Statistical Confidence Limits Technique

The use of image segmentation in image processing is of great importance in analyzing and extracting information from images, and one of the most important segmentation techniques is the threshold technique, which is considered one of the simplest techniques of image division in image processing. The statistical methods play an important role in the process of image segmentation. Statistical confidence in image processing, preliminary processing, as it removed noise from the images, and here the obscure noise was used. After that, the resulting images were cut, the initial processing process with the global Otsu threshold technology and a group of local techniques, namely Niblack, sauvola and local Bernsen, and the split image quality was measured by statistic measures namely Jaccard Similarity Coefficient and Maximum Signal to Noise Ratio (PSNR). as was the application of the methods mentioned on the images and the comparison between the methods of treatment in order to obtain the best results that appear in the image in which it appears and reduce noise.

Deepfake в исследованиях восприятия лица

Presents the state-of-the-art Deepfake face replacement image collage method, an artificial intelligence (AI) product that can be used to create high-quality, realistic videos with a fake or replaced face, with no obvious signs of manipulation. Based on the DeepFaceLab (DFL) application, the process of creating video images of an “impossible face” is described step by step. The results of the experiments of studying the perception patterns of the moving “impossible face” and their differences in statics and dynamics are presented. The stimuli were two DFL-generated models of virtual sitters with impossible faces: a video image of a chimerical face, in which the right and left sides belong to different people, and a Tatchered face with the eyes and mouth areas rotated by 180°. It was shown that the phenomena of perception of the “impossible face”, registered earlier under static conditions (integrity of perception of the split image, distraction and inversion effect), are preserved and acquire a new content when dynamic models are exposed. In contrast to the collaged images, the original faces in statics and motion, regardless of egocentric orientation, are evaluated positively at the level of high values. Under all tested conditions the gender of the virtual sitter is determined adequately, the perceived age is overestimated. Estimates of the virtual sitter’s emotions from his video images are differentiated into basic (stable) and additional (changing) states, the ratio of which depends on the content of a particular episode. Deepfake image synthesis technology significantly expands the possibilities of psychological research of interpersonal perception. The use of digital technologies simplifies the creation of “impossible face” stimulus models necessary for in-depth study of representations of the human inner world, and creates a need for new experimental-psychological procedures corresponding to a higher level of ecological and social validity.

Paisagem desbotada nos (des)caminhos da vida: uma leitura de Paisagem com mulher e mar ao fundo, de Teolinda Gersão

A memória exerce papel fundamental na construção da subjetividade, conserva aquilo que, de alguma forma, ficou marcado na trajetória de vida. Consequentemente, os espaços nos quais o sujeito vivencia experiências transformam-se em espaços de memória, já que neles ficam registradas impressões e são capazes de absorver vivências acumuladas. Dessa forma, o trabalho em questão propõe analisar na obra Paisagem com mulher e mar ao fundo de Teolinda Gersão (1982), a relação que a protagonista estabelece com o espaço e a paisagem, e os sentidos que ganham nas suas recordações. A paisagem anunciada no título traduz o tom imagético e acústico que envolve a cena central da narrativa, numa relação simbiótica de presença/ausência. As lembranças da protagonista voltam-se para os espaços de referência, repletos de imagens cindidas, as quais reposicionam o espaço da casa e a paisagem da aldeia e do mar, realçados pelo desejo de realocação de cenas familiares.

An algorithm for the split feasible problem and image restoration

This paper proposes an accelerated algorithm for the split common fixed point problem, based on viscosity approximation methods and inertial effects. The main result will be applied to image restoration problems. This algorithm is constructed in such a way that its step sizes and the norm of a given linear operator are not related. Under some conditions, the strong convergence of the algorithm is obtained. Numerical investigations are carried out in order to illustrate high-performance of the present work, mainly using processing duration and the signal-to-noise ratio. It is also shown that this proposed algorithm is more efficient and effective than the published algorithm by Yao et al.

Three-Dimensional Reconstruction of Fleece Fabric Surface for Thickness Evaluation

Aiming at solving the problem of manually measuring the fabric surface thickness, this paper proposes a three-dimensional (3D) reconstruction method based on the tangential two-dimensional (2D) sequence images. Firstly, the characteristic region of the fabric surface is extracted. Secondly, the image is splitting based on the maximum between-class variance method. Thirdly, the splitting image is processed by the morphological method. Fourthly, the canny operator is used to obtain the edge detection for calculating the edge contour coordinate. Finally, the surf function is used to reconstruct the 3D model of the fabric surface. To evaluate the performance of the proposed 3D model, the thickness and relief degree of the fabric surface are used, and the comparison between the proposed method and the manual measurement is carried out. The results demonstrate that, under a reasonable relief degree condition, the proposed method is more effective to evaluate the thickness of the fabric surface and the estimated thickness is more accurate than the manually measured one.

A Single Image Mosquito Streaks Removal Method of High-Speed Railway Traction Substation Using DCNN

With the proposal of intelligent high-speed railway, the research on remotely monitoring the intelligent traction substation becomes a key subject of high-speed railway safe operation. However, mosquitos adhered to the glass window or camera lens can severely hamper the visibility of a background scene, and degrade images considerably. Therefore, we propose a single image mosquito streaks removal method of high-speed railway traction substation based on the deep convolutional neural network. First, we employ guided filter to split input image into smoothing image and edge-preserving image. Then, the edge-preserving image is fed into our designed convolutional neural network to obtain learning map, which solves problems of background interference and focuses the model on the structure of rain streak in images, and the clean image is finally generated through adding the input image and learning map. The experiment results on Heishan traction substation real datasets show the effectiveness of our proposed method.

Multi-modal medical image fusion by Laplacian pyramid and adaptive sparse representation

Multi-modal medical image fusion refers to the fusion of two or more medical images obtained by different imaging methods into one image. Multi-modal medical images contain a lot of useful information that helps doctors to make a diagnosis. In this study, a multi-modal medical image fusion method is proposed based on Laplacian pyramid (LP) decomposition and adaptive sparse representation (ASR). ASR was used to reduce the noise of high-frequency information in the image fusion process and it did not need a high redundancy dictionary as traditional sparse representation (SR) methods. The proposed fusion method first used the LP decomposition to split medical images into four images of different sizes. Then ASR was performed to fuse the decomposed four layers, respectively. Finally, the fused image was obtained by the inverse Laplace pyramid transform. Experimental results showed that the proposed method could effectively fuse the medical images with the detailed information perfectly integrated, and could also reduce the influences of artifacts, noise and block effect. The research results are of great significance in the field of medical image fusion.