Vasculature feature segmentation from the fundus images is critical for the identification of retinal diseases. However, automated vessel segmentation is challenging owing to variability in vessel structure and color gradients, low contrast between vessels, and pathologies. The state-of-the-art approaches address the challenges of emerging hand-crafted filters to apprehend vessel-like patterns. More recently, deep learning-based methods have evolved for the vessel segmentation task, which employs annotations to train the model. These approaches ignore the vessel's geometrical characteristics in the fundus images, leading to inaccuracies. Herein, we propose a novel unsupervised segmentation method based on interrelationship handling, Bonferroni mean pre-aggregation operator, with the aid of dynamic fuzzy histogram equalization, namely BMPDFHESeg. The method extracts the vascular information by fusing color channels by constructing an interrelationship handling pre-aggregation operator. The operator enables finding the direction of increasingness to segment large vessels and vessel feature enhancement through a dynamic fuzzy histogram equalization process using the prior feature intensity information. BMPDFHESeg is assessed qualitatively and quantitatively using the DRIVE, STARE, and HRF datasets, demonstrating enhanced efficacy and computational speed. Further, the results were validated by ophthalmologists for the accuracy of the vessel segmentation and usefulness for the diagnosis of retinal disorders.

In recent days, there’s been a rise in billing fraud, including invoice fraud, credit card fraud, and online payment fraud, with fraudsters using various tactics to trick individuals and businesses. While the security methods are often fails to address high-speed digital forgery techniques. To address this limitation in this research we present an advanced approach for detecting tampering and preventing forgery in customers retail billing and receipts using the AI tool such as convolutional neural networks (CNN), updated sand cat swarm optimization (USCSO) and gray-level co-occurrence matrix (GLCM). The utilization of the CNN hyperparameter tuning with USCSO, that outperforming standard optimizations like Adam and SGD. With the use of feature extraction, classification, and image preparation (such as skeletonization, color conversion, and scaling), the approach can accurately differentiate between authentic and fraudulent documents. The model’s resilience to various document type and forgery tactics is confirmed by extensive testing on the SROIE dataset and a private dataset of 5000 images. A CNN, optimized by USCSO, classifies these features, achieving 97.1% accuracy, 97.3% precision, and 97.5% recall in detecting fake text, outperforming traditional CNN and SVM techniques.

Bright-field microscopy (BFM) encrypts the optical transillumination profile of the transmitted light attenuated by the complex micro-structural tissue convolutions, manifested by the dense and compact regions of the specimen under examination. The connotations of idiosyncratic tissue interaction dynamics with the onset of pre-cancerous activity are encoded in the BFM acquired oral mucosa histopathological images (OMHI). In the present study, our analysis is focused on the sub-epithelium region of the oral mucosa, which has high clinical significance but sparsely explored in the literature from the textural domain. Histopathology being the gold-standard technique till date, we have used the light microscopic histopathology images for tissue characterization. The tissue-index transmission patches (TITP) from the sub-epithelium region are cropped under the guidance of oral onco-pathologists. After that, the TITPs are characterized for its multi-scale spatial-deformation dynamics, while keeping the intrinsic anisotropic geometry, and local contour connectivity within tolerable limits. With recent studies exhibiting multifractal's potency in diverse biological system analysis, here, we exploit the 2D multifractal detrended fluctuation analysis (2D-MFDFA) on TITPs for exploring a discriminative set of multifractal signatures for healthy, oral potentially malignant disorders and oral cancer tissue sample. The predictive model's competency is validated on an experimentally collected corpus of TITP samples and substantiated via confirmatory data statistics and analysis, showing its inter-class segregation efficacy. Moreover, the 2D-MFDFA analysis evinces the complex multifractal patterns in TITPs, which is due to the presence of composite long-range correlations in the oral mucosa tissue fabric.

Fuzzy rule interpolation (FRI) predicts an accountable outcome of a possible course of action in sparse fuzzy rule base system (FRBS). However, in real life, we encounter some situations where the antecedent has to be predicted to obtain a desired consequent of FRBS. In this situation, inverse fuzzy rule interpolation (IFRI) or backward fuzzy rule interpolation (BFRI) is used to get the desired outcome. Here a geometry based inverse fuzzy rule base interpolation (GIFRI) is suggested. The mathematical detail of the proposed method is elaborated and its geometrical interpretation is given with the help of fuzzy geometry. It is to be noted that the proposed method ensures that the inverse of the inverse is the original one.

The concept of pre-aggregation functions, which was oriented as an elementary attempt to outstretch the notion of monotonicity in aggregation functions, has enlarged the class of operators for information accumulation by considering directional monotonicity with respect to a vector. This consideration propels us to focus on the systematic investigation of the theoretical framework of different forms of pre-aggregation functions, particularly Bonferroni mean-type. In this regard, we propose the construction methodology of Bonferroni mean-type (BM-type) pre-aggregation functions by befitting suitable functions to provide a descriptive configuration, which is quite interpretable and understandable. Firstly, a construction method of BM-type pre-aggregation function has been propounded by utilizing a bivariate function M. Its properties are inspected in detail. To enrich its capacity, the proposed BM-type pre-aggregation function has been customized by utilizing two functions, namely M and M*, respectively. Several illustrative examples have been presented in this regard.

The partitioned Bonferroni mean (PBM) operator, which was oriented as an elementary attempt to outstretch the Bonferroni mean (BM) operator, has enlarged the class of BM-type aggregation operators for information accumulation by modeling interrelationship among pairwise disjoint partition sets with the presupposition that the criteria of intra-partition are homogeneously related to each other, while no relationship exists among criteria of inter-partition. Although PBM has encountered a lot of attraction from the researchers due to its versatility in information aggregation technique, the principal disadvantage of the existing PBM definitions evolution is that they do not provide any specification regarding the relationship among criteria of partition structure during design, development, and applications of PBM over unalike situations of information fusion. This consideration propels us to focus on the systematic investigation of different variations of PBM operators based on various mandatory requisites to be imposed on information retrieved from the partition sets. In this regard, we propose the construction of novel generalized partitioned Bonferroni mean (GPBM) operator by befitting its suitable components to provide a descriptive configuration, which is quite interpretable, understandable and thus facilitates the ability to model specific mandatory prerequisites in a single operator. To enrich the capacity for modeling real-life decision situations, the PBM operator is customized to propose optional partitioned Bonferroni mean (OPBM) operator that captures partition-wise interrelationship among attributes while taking into consideration optional conditions jumbled in each partition set. Furthermore, we demonstrate the construction methodology of generalized OPBM operator that amalgamate the concept of GPBM and OPBM operator to enhance and model-specific requirements along with optional requirements as per the desires of decision makers.

In this paper, a fuzzy multi-objective mathematical programming problem is considered where functional relationship between the decision variables and the objective functions is not completely known to us. It is assumed that the information source from where some knowledge may be obtained about objective functions consists of a block of fuzzy if-then rules. The focus in this work is to solve the multi-objective optimization problem for the above situations. A numerical example is also presented to illustrate the method.

This paper is the first which attempts to capture complete fuzzy non-dominated set of fuzzy multi-criteria optimization problems with fuzzy parameters. A proper mathematical formulation of fuzzy non-dominated set and its generation are primary aims of the proposed study. Present work is mainly focused on visualizing the considered problem from fuzzy geometrical viewpoint. Constraint set of a fuzzy multi-criteria optimization problem is viewed from two different spaces - decision space and criterion space. Fuzzy decision feasible region or the constraint set on decision space is formulated using the alpha-cuts of the parameters present in the problem. Under the assumption that fuzzy criteria are fuzzy number valued, it is shown that a fuzzy point will be obtained on the criterion space corresponding to each point on the decision feasible region. Union of all these fuzzy points determines fuzzy criteria feasible region or constraint set in the criterion space. To capture entire fuzzy non-dominated set of criteria feasible region, a method, hereby named fuzzy ideal cone method, has been proposed. The method essentially uses the cone of non-positive hyperoctant of the criteria space to generate complete fuzzy non-dominated set. Proposed methodology is supported by several numerical examples and pictorial illustrations.

In developing so called fuzzy expert systems, fuzzy rule bases have been considered with greater importance. In fact, a fuzzy rule base is a knowledgebase that models human cognitive factors. Fuzzy rules are linguistic ‘IF-THEN’ constructions where ‘IF’ part consists of a set of fuzzy variables and ‘THEN’ part includes a dependent fuzzy variable. In order to identify the underlying mathematical structure in the fuzzy rule base, we develop fuzzy linear and fuzzy polynomial regression techniques in this paper. And the estimation of model parameters is also shown using least-square approach. Finally, examples are illustrated to demonstrate the proposed model.