Article
Computer Science, Information Systems
Chunzhi Wang, Chengkun Tu, Siwei Wei, Lingyu Yan, Feifei Wei
Summary: This paper proposes a multilevel thresholding image segmentation technique based on an improved whale optimization algorithm. The results of algorithm evaluation experiments demonstrate that the MSWOA has higher search accuracy and faster convergence speed. The image segmentation experimental results show that the MSWOA-Kapur technique can effectively and accurately search multilevel thresholds.
Article
Multidisciplinary Sciences
Antony Onim, S. Musyoki, P. Kihato
Summary: Interference is a major challenge in cellular networks, and Fractional Frequency Re-use is a method used to mitigate it. By dividing cells into partial re-use and full re-use regions and setting a Signal-to-Interference-plus-Noise-Ratio threshold, bandwidth can be allocated based on user numbers. This study develops a dynamic thresholding technique and compares its performance with other methods.
Article
Physics, Multidisciplinary
Qingyu Deng, Zeyi Shi, Congjie Ou
Summary: A self-adaptive segmentation algorithm is proposed in this study, combining the advantages of entropy-based and variance-based image thresholding methods to deal with a wider range of images. The new algorithm shows better performance in correctness and robustness compared to other famous image thresholding algorithms, as demonstrated by four quality indices. Additionally, the algorithm has potential application in self-adaptive object recognition.
Article
Computer Science, Artificial Intelligence
Essam H. Houssein, Bahaa El-din Helmy, Diego Oliva, Ahmed A. Elngar, Hassan Shaban
Summary: This paper introduces the use of the Black Widow Optimization (BWO) algorithm to find the best threshold configuration for multi-level image segmentation, achieving higher efficiency and reliability compared to other meta-heuristic algorithms. Experimental results demonstrate the superior performance of the BWO-based method, showing potential applications in the field of image processing.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Mathematical & Computational Biology
Shikai Wang, Kangjian Sun, Wanying Zhang, Heming Jia
Summary: The paper proposes a modified ant lion optimizer algorithm based on opposition-based learning for optimizing multilevel thresholding in image segmentation, and experimental results show that the method outperforms others in terms of segmentation performance.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Jiquan Wang, Jinling Bei, Haohao Song, Hongyu Zhang, Panli Zhang
Summary: This paper proposes a whale optimization algorithm with combined mutation and removing similarity (CRWOA) to address the issues of weak local search ability and imbalance between exploration and exploitation in the traditional whale optimization algorithm (WOA). The CRWOA introduces an adaptive adjustment method, a roulette selection operator and a combined mutation operator to enhance the algorithm's performance. Experimental results demonstrate the superiority of CRWOA in terms of convergence and segmentation quality.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Hardware & Architecture
Longzhen Duan, Shuqing Yang, Dongbo Zhang
Summary: An improved cuckoo search algorithm (ICS) is proposed for multilevel thresholding image segmentation in this paper, and the experimental results show that the proposed algorithm is superior to other seven well-known heuristic algorithms.
JOURNAL OF SUPERCOMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Essam H. Houssein, Kashif Hussain, Laith Abualigah, Mohamed Abd Elaziz, Waleed Alomoush, Gaurav Dhiman, Youcef Djenouri, Erik Cuevas
Summary: The paper introduces an enhanced version of the Marine Predators Algorithm (MPA) called MPA-OBL, which incorporates Opposition-Based Learning (OBL) to improve search efficiency and convergence. Through comprehensive experiments, MPA-OBL is shown to outperform other algorithms in solving complex optimization problems, demonstrating superior quality of solutions and faster convergence speed.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Engineering, Multidisciplinary
R. Srikanth, K. Bikshalu
Summary: Image segmentation is the process of dividing an image into regions, with multilevel thresholding being a mature method and Otsu's method being a significant technique. Traditional histogram-based methods lack spatial details of contextual information, leading to the proposal of a novel method using Energy Curve instead.
AIN SHAMS ENGINEERING JOURNAL
(2021)
Article
Computer Science, Information Systems
Simrandeep Singh, Nitin Mittal, Harbinder Singh, Diego Oliva
Summary: Image segmentation is a critical stage in image analysis and pre-processing, where pixels are divided into segments based on threshold values. Multi-level thresholding approaches are more effective than bi-level methods, and a new modified Otsu function is proposed that combines Otsu's between-class variance and Kapur's entropy. Experimental results demonstrate the high efficiency of the modified Otsu method in terms of performance metrics.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Manoj Kumar Naik, Rutuparna Panda, Ajith Abraham
Summary: The study introduces a context-sensitive entropy dependency-based multilevel thresholding method, accompanied by the opposition equilibrium optimizer. Through various testing and analysis, the method is shown to demonstrate advantages in reducing complexity, improving accuracy, and enhancing stability and scalability.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Information Systems
Aneesh Wunnava, Manoj Kumar Naik, Rutuparna Panda, Bibekananda Jena, Ajith Abraham
Summary: This study introduces an improved Harris hawks optimization algorithm, DEAHHO, which incorporates differential evolution and adaptive exploration to enhance the performance of search agents, successfully applied in image thresholding.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Automation & Control Systems
Bibekananda Jena, Manoj Kumar Naik, Rutuparna Panda, Ajith Abraham
Summary: This study proposes a multilevel thresholding method based on 3D Tsallis entropy, which is more robust than the 1D/2D Tsallis entropy methods. The introduction of the attacking Manta Ray foraging optimization algorithm shows superior performance compared to state-of-the-art optimization algorithms, making it useful for multi-spectral color image segmentation.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Environmental Sciences
Shreya Pare, Himanshu Mittal, Mohammad Sajid, Jagdish Chand Bansal, Amit Saxena, Tony Jan, Witold Pedrycz, Mukesh Prasad
Summary: Segmentation techniques in remote sensing imagery face challenges such as dense features, low illumination, uncertainties, and noise. Existing multilevel thresholding methods lack spatial information, leading to low segmentation accuracy.
Article
Computer Science, Information Systems
Taymaz Rahkar Farshi, Ahad K. Ardabili
Summary: This study explores optimizing image segmentation with a hybrid optimization algorithm and applying it to multilevel image thresholding. The experiments show that the proposed method outperforms other optimization algorithms in various metrics.
MULTIMEDIA SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Reman Kumar, Ashish Kumar Bhandari
Summary: This article proposes a computationally efficient scaled clustering-based contrast enhancement method to improve image quality while maintaining color and naturalness. The method groups pixels into clusters using the fuzzy c-means algorithm and modifies pixel intensity based on the modified and scaled cluster memberships. The enhanced image achieved through linear scaling shows better contrast and color information without artifacts.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Kankanala Srinivas, Ashish Kumar Bhandari
Summary: This article presents a novel context-based histogram bin stretching method called CHBS, which enhances contrast by increasing the range of gray levels and randomness among the gray levels. It combines spatial contextual information and discrete cosine transform (DCT) to achieve global enhancement and local detail enhancement. The method first generates random numbers based on the spatial similarities among surrounding pixels, and then distributes intensity values among the available dynamic range to generate a globally contrast-enhanced image. The DCT technique is further applied to adjust the local details automatically. Experimental results demonstrate that the proposed approach outperforms or is comparable to state-of-the-art methods in terms of brightness preservation, richer details, and natural appearance.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Review
Computer Science, Interdisciplinary Applications
Sanat Kumar Pandey, Ashish Kumar Bhandari
Summary: Lung cancer is a prevalent form of cancer that can affect individuals of all age groups. Detecting lung cancer in its early stages is crucial for controlling cell growth and minimizing mortality rates. However, the complex structure of the human lung poses challenges for radiologists in detecting cancerous regions. This manuscript reviews various smart healthcare models based on artificial intelligence, specifically machine learning and deep learning, for lung cancer detection and classification. It analyzes research literature, including deep learning and machine learning models with high accuracy and efficiency, published from 2012 to 2022. The report also identifies research gaps and advancements in technology for effective detection of cancerous parts of the human lung, aiming to assist experts in developing more efficient models for lung cancer diagnosis and inspiring new researchers to contribute to healthcare advancements.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Computer Science, Information Systems
Neha Singh, Ashish Kumar Bhandari
Summary: A novel multilevel image segmentation framework is proposed, using variational mode decomposition (VMD) and multiclass variance function for segmentation. The unfavorable effects of histogram fluctuation are eliminated by decomposing it into sub-modes for analysis and attribute extraction. The Otsu function is then utilized to generate accurate and optimal threshold values for desired image segmentation. Experimental results demonstrate that the proposed technique produces improved segmented images compared to other state-of-the-art techniques.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Anshuman Prakash, Ashish Kumar Bhandari
Summary: This paper presents a novel adaptive image enhancement technique, Cuckoo Search Constrained Gamma Masking, for improving the quality of MRI images and providing computational support.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Review
Computer Science, Interdisciplinary Applications
Sushi Sushanki, Ashish Kumar Bhandari, Amit Kumar Singh
Summary: Breast cancer is a common type of cancer that requires early detection. Recent advancements in multi-modal imaging and machine learning algorithms have shown potential in improving the accuracy of breast cancer diagnosis, providing crucial information for physicians.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Review
Computer Science, Interdisciplinary Applications
Poonam Rani Verma, Ashish Kumar Bhandari
Summary: Image classification is the labeling of pixels or voxels in an image based on certain rules. It has applications in medical image analysis, satellite image identification, etc. In medical images, segmentation is often performed to extract necessary areas, followed by classification based on certain criteria. This is important for detecting disorders and studying specific human organs in detail.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Review
Computer Science, Interdisciplinary Applications
Preity, Ashish Kumar Bhandari, Syed Shahnawazuddin
Summary: Ocular diseases are on the rise, causing vision loss at an early age, with cardiovascular diseases being the primary reason. Analyzing the internal structure of the retina, particularly vasculature, is crucial for early diagnosis. Researchers have proposed various methods for vessel segmentation and disease detection, constantly improving performance parameters. This paper presents a systematic review of existing methodologies and compares different approaches for computer-aided diagnosis of ocular diseases.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Computer Science, Information Systems
Manish Kumar, Ashish Kumar Bhandari, Manvi Jha
Summary: Images are crucial in various scientific applications, but their quality and visibility are greatly affected by environmental and lighting conditions. Existing methods for addressing visibility issues in low-illumination and uneven illumination scenes have been proven unsatisfactory. This research proposes an effective approach that uses a brightness transfer function based on the Weber-Fechner law to enhance dark images with uneven illumination. The proposed function is adaptively controlled using the value and saturation components of the input image in HSV space. Image fusion technique and a novel color contrast enhancement function based on the H-K effect are also employed. Qualitative and quantitative assessments confirm the superiority of the proposed approach over existing methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Sidharth Kashyap, Ashish Kumar Bhandari, Pushpa Giri
Summary: This paper presents a resource-efficient architecture for edge detection techniques (Sobel, Prewitt, Roberts, and Laplacian) on FPGAs. The proposed architecture is simpler and uses fewer resources compared to conventional methods. It achieves reduced arithmetic operations and improved power consumption by implementing convolution kernel properties, pipelining, and concurrent output methods. The architecture also includes a buffering scheme to decrease the use of Block RAM and bandwidth between memory and FPGA. The Sobel edge detector with the proposed architecture shows superior performance compared to other methods and can process 325 frames per second with a power consumption of 0.107 W.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Sanat Kumar Pandey, Ashish Kumar Bhandari
Summary: This paper proposes a deep learning-based transfer learning technique using YOLOv5 for efficient and accurate detection of brain tumors. Through comparisons with other deep learning models, it is found that the YOLOv5 model performs the best. The model was trained using the BraTS21 dataset and achieved high accuracy in the experiments.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ashwini Kumar Upadhyay, Ashish Kumar Bhandari
Summary: By making architectural changes and modifications in the training process, we propose a modified U-Net (Mod-UNet) model for automatic lung infection segmentation. The results on two Covid-19 Lung CT segmentation datasets show that Mod-UNet outperforms the baseline U-Net model. Furthermore, to address the issue of lack of annotated data, we introduce a semi-supervised framework (Semi-Mod-UNet) that significantly improves the segmentation performance.
IEEE TRANSACTIONS ON RADIATION AND PLASMA MEDICAL SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Bharath Subramani, Magudeeswaran Veluchamy, Ashish Kumar Bhandari
Summary: Contrast enhancement is crucial in medical imaging technology for clinical diagnosis, but degraded medical images pose challenges for disease diagnosis. This article presents an improved fuzzy transformation model to enhance contrast by adjusting image intensity levels. The proposed method, optimized with moth flame optimization, expands the dynamic range and improves visibility of local details in medical images.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Rakesh Ranjan, Bikash Chandra Sahana, Ashish Kumar Bhandari
Summary: Sleep is crucial for human well-being, and sleep EEG signals need to be assessed for diagnosing sleep-related neurological disorders. A hybrid signal denoising method has been proposed for automatic detection and suppression of cardiac artifacts from single-channel EEG, which outperforms other automated EEG artifact elimination methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Reman Kumar, Ashish Kumar Bhandari, Manish Kumar
Summary: In this article, an effective contrast enhancement algorithm is proposed to enhance the contrast of an image while preserving color details and maintaining visual similarity to the undistorted reference image. The algorithm is divided into two parts, color saturation enhancement and brightness enhancement. A haze removal approach using the atmospheric scattering model is used to formulate a color saturation enhancement function, and the brightness of the enhanced output is restored using the proposed enhancement function. The proposed algorithm successfully preserves the natural appearance and brightness of the image.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)