Debashis Nandi

Bio: Debashis Nandi is an academic researcher from National Institute of Technology, Durgapur. The author has contributed to research in topics: Encryption & Image segmentation. The author has an hindex of 10, co-authored 66 publications receiving 415 citations. Previous affiliations of Debashis Nandi include Techno India & Indian Institute of Technology Kharagpur.

An Image Encryption Process based on Chaotic Logistic Map

Abstract: This paper proposes a high security image encryption technique using logistic map. The proposed image encryption algorithm is described in detail along with its security analysis such as key space analysis, statistical analysis and differential analysis. A comparison in terms of correlation between the initial and transformed images, Number of pixels change rate and unified average changing intensity is also done. The present algorithm has been tested using different images to prove that the encryption method has a great potential and has a good ability to achieve the high confidential security.

Optimal location of UPFC controller in transmission network using hybrid chemical reaction optimization algorithm

Abstract: This paper presents a new hybrid chemical reaction optimization (HCRO) approach which is based on chemical reaction optimization (CRO) and differential evolution (DE) to find the optimal placement and parameter setting of unified power flow controller (UPFC) to achieve optimal performance of power system network. In the proposed algorithm, four elementary reactions, i.e., on-wall ineffective collision, inter-molecular ineffective collision, decomposition, and synthesis, are developed. Moreover, mutation operation of DE is integrated with inter-molecular ineffective collision and crossover operation is introduced in the inter-molecular collision, synthesis, and decomposition process to accelerate the convergence speed and improve the solution quality of CRO algorithm. Here, three different single objectives namely, minimization of the overall cost, transmission loss, voltage deviation and one multi-objective which simultaneously minimizes the transmission loss and voltage deviation are used. To verify the effectiveness, the proposed HCRO approach is implemented on IEEE 14-bus and IEEE 30-bus power systems. Moreover, to establish the superiority, the simulation results of the proposed HCRO technique are compared to the CRO and other previously reported algorithms published in the literature such as genetic algorithm (GA), particle swarm optimization (PSO), immune GA (IGA), immune PSO (IPSO) and hybrid immune algorithm (HIA). It is found that the results obtained by the proposed HCRO technique are superior to those obtained by other discussed algorithms.

Brain Tumor Classification Using ResNet-101 Based Squeeze and Excitation Deep Neural Network

Abstract: The brain tumor is one of the leading and most alarming cause of death with a high socio-economic impact in Occidental as well as eastern countries. Differential diagnosis and classification of tumor types (Gliomas, Meningioma, and Pituitary tumor) from MRI data are required to assist radiologists as well as to avoid the dangerous histological biopsies. In the meantime, improving the accuracy and stability of diagnosis is also one challenging task. Many methods have been proposed for this purpose till now. In this work, an automatic tool for classification of brain tumor from MRI data is presented where the image slice samples are passed into a Squeeze and Excitation ResNet model based on Convolutional Neural Network (CNN). The use of zero-centering and normalization of intensity for smooth variation of the intensity over the tissues was also investigated as a preprocessing step which together with data augmentation proved to be very effective. A relative study had been done to prove the efficacy of the proposed CNN model in free tumor database. Experimental evaluation shows that the proposed CNN archives an overall accuracy rate of 89.93% without data augmentation. Addition of data augmentation has further improved the accuracies up to 98.67%, 91.81% and 91.03% for Glioma, Meningioma and Pituitary tumor respectively with an overall accuracy of 93.83%. Promising improvement with reference to sensitivity and specificity compared with some of the state-of-the-art methods was also observed.

Optimal location of STATCOM using chemical reaction optimization for reactive power dispatch problem

Abstract: Optimal reactive power dispatch (ORPD) problem has a significant influence on optimal operation of power systems. However, getting optimal solution of ORPD problem is a strenuous task for the researchers. The inclusion of flexible AC transmission system (FACTS) devices in the power system network for solving ORPD problem adds to its complexity. This paper presents the application of chemical reaction optimization (CRO) for optimal allocation of a static synchronous compensator (STATCOM) to minimize the transmission loss, improve the voltage profile and voltage stability in a power system. The proposed approach is carried out on IEEE 30-bus and IEEE 57-bus test systems and the simulation results are presented to validate the effectiveness of the proposed method. The results show that the proposed approach can converge to the optimum solution and obtains better solutions as compared to other methods reported in the literature.

Unified power flow controller based reactive power dispatch using oppositional krill herd algorithm

Abstract: In power system, minimizing the power loss in the transmission lines and/or minimizing the voltage deviation at the load buses by controlling the reactive power is referred as optimal reactive power dispatch (ORPD). This paper presents an improved evolutionary algorithm based on oppositional krill herd algorithm (OKHA) for obtaining optimal steady-state performance of power systems. This article also proposes the effect of UPFC location in steady-state analysis and to demonstrate the capabilities of UPFC in controlling active and reactive power flow within any electrical network. To verify the effectiveness of KHA and OKHA, two different single objective functions such as minimization of real power losses and improvement of voltage profile and a multi-objective function that simultaneously minimizes transmission loss and voltage deviation have been studied through standard IEEE 57-bus and 118-bus test systems and their results have been reported. The study results show that the proposed KHA and OKHA approaches are feasible and efficient.

A comprehensive review

Abstract: Graphical abstractDisplay Omitted HighlightsThe comprehensive review of Krill Herd Algorithm as applied to different domain is presented.The review covers the applications, modifications and hybridizations of the KH algorithms.It provides future research directions across different areas. Krill Herd (KH) algorithm is a class of nature-inspired algorithm, which simulates the herding behavior of krill individuals. It has been successfully utilized to tackle many optimization problems in different domains and found to be very efficient. As a result, the studies has expanded significantly in the last 3 years. This paper presents the extensive (not exhaustive) review of KH algorithm in the area of applications, modifications, and hybridizations across these fields. The description of how KH algorithm was used in the approaches for solving these kinds of problems and further research directions are also discussed.

U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications

Abstract: U-net is an image segmentation technique developed primarily for image segmentation tasks. These traits provide U-net with a high utility within the medical imaging community and have resulted in extensive adoption of U-net as the primary tool for segmentation tasks in medical imaging. The success of U-net is evident in its widespread use in nearly all major image modalities, from CT scans and MRI to X-rays and microscopy. Furthermore, while U-net is largely a segmentation tool, there have been instances of the use of U-net in other applications. Given that U-net’s potential is still increasing, this narrative literature review examines the numerous developments and breakthroughs in the U-net architecture and provides observations on recent trends. We also discuss the many innovations that have advanced in deep learning and discuss how these tools facilitate U-net. In addition, we review the different image modalities and application areas that have been enhanced by U-net.

A novel image encryption algorithm based on a 3D chaotic map

Abstract: Recently [Solak E, Cokal C, Yildiz OT Biyikogˇlu T. Cryptanalysis of Fridrich’s chaotic image encryption. Int J Bifur Chaos 2010;20:1405–1413] cryptanalyzed the chaotic image encryption algorithm of [Fridrich J. Symmetric ciphers based on two-dimensional chaotic maps. Int J Bifur Chaos 1998;8(6):1259–1284], which was considered a benchmark for measuring security of many image encryption algorithms. This attack can also be applied to other encryption algorithms that have a structure similar to Fridrich’s algorithm, such as that of [Chen G, Mao Y, Chui, C. A symmetric image encryption scheme based on 3D chaotic cat maps. Chaos Soliton Fract 2004;21:749–761]. In this paper, we suggest a novel image encryption algorithm based on a three dimensional (3D) chaotic map that can defeat the aforementioned attack among other existing attacks. The design of the proposed algorithm is simple and efficient, and based on three phases which provide the necessary properties for a secure image encryption algorithm including the confusion and diffusion properties. In phase I, the image pixels are shuffled according to a search rule based on the 3D chaotic map. In phases II and III, 3D chaotic maps are used to scramble shuffled pixels through mixing and masking rules, respectively. Simulation results show that the suggested algorithm satisfies the required performance tests such as high level security, large key space and acceptable encryption speed. These characteristics make it a suitable candidate for use in cryptographic applications.

Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem

Abstract: Automated segmentation of anatomical structures is a crucial step in image analysis. For lung segmentation in computed tomography, a variety of approaches exists, involving sophisticated pipelines trained and validated on different datasets. However, the clinical applicability of these approaches across diseases remains limited. We compared four generic deep learning approaches trained on various datasets and two readily available lung segmentation algorithms. We performed evaluation on routine imaging data with more than six different disease patterns and three published data sets. Using different deep learning approaches, mean Dice similarity coefficients (DSCs) on test datasets varied not over 0.02. When trained on a diverse routine dataset (n = 36), a standard approach (U-net) yields a higher DSC (0.97 ± 0.05) compared to training on public datasets such as the Lung Tissue Research Consortium (0.94 ± 0.13, p = 0.024) or Anatomy 3 (0.92 ± 0.15, p = 0.001). Trained on routine data (n = 231) covering multiple diseases, U-net compared to reference methods yields a DSC of 0.98 ± 0.03 versus 0.94 ± 0.12 (p = 0.024). The accuracy and reliability of lung segmentation algorithms on demanding cases primarily relies on the diversity of the training data, highlighting the importance of data diversity compared to model choice. Efforts in developing new datasets and providing trained models to the public are critical. By releasing the trained model under General Public License 3.0, we aim to foster research on lung diseases by providing a readily available tool for segmentation of pathological lungs.