Search

Photovoltaic systems have become more popular as people become more interested in developing energy from renewable resources. Even after the installations, however, there is still a lack of understanding about the importance of inspecting the condition of the PV modules. To keep the PV running, early hot-spot detection is required. For detecting hot-spots, thermal imaging is still a popular technique. This research proposes to develop a method for detecting hot-spots in thermal images of photovoltaic modules using artificial intelligence techniques. Pre-processing, segmentation with an Artificial Neural Network and identification are the three main processes. The proposed method appears to be a good choice for improving hot-spot detection in PV monitoring systems, according to the results of the experiments.

In photovoltaic systems, maximum power point tracking (MPPT) methods are used to get the maximum power out of them. The presence of several peaks in a PV array’s power-voltage characteristics is due to partial shadowing conditions that increase the complexity of the tracking operation. In this paper, the Global Maximum Power Point (GMPP) is calculated using latest meta-heuristic optimization algorithm known as Jellyfish Search (JS). This unique MPPT approach is used to reduce PV module tracking time and improve the tracking efficiency. Using MATLAB/SIMULINK, the effectiveness of the suggested JS algorithm is assessed by contrasting it with the traditional P&O approach in terms of tracking speed and precision. The simulation findings indicate that the JS algorithm’s tracking ability is better than that of the conventional P&O method. In the experimental results, the JS algorithm reduces convergence time by 56.6% when compared to the PSO algorithm. Also, the proposed JS algorithm generates output power higher than the PSO MPPT algorithm using the duty cycle ratio value at the expected peaks.

Diabetic Macula Edema (DME) is brought about by collection of extracellular liquid from hyper penetrable vessels inside the macula. DME is one of the main sources of visual impairment among diabetic patients. The patients who are influenced by diabetes and Diabetic Retinopathy (DR) are well on the way to be influenced by DME. The presence of Exudates in macula area is the significant sign of essence of DME. The Optic Disc (OD) and Blood Vessels (BVs) are the retinal milestones divided to find macula and fovea locale. Area of macula focus is vital for portion the anomalies in the macula locale. The irregularity and its seriousness of the sickness in macula district can be distinguished solely after finding the macula place. Pictures from High Resolution Fundus (HRF) database are utilized for testing and examination. The Atanassov’s Intuitionistic fuzzy Sets (A-FIS) and morphological tasks are utilized to portion OD and BVs. With these retinal landmarks, macula is recognized by raphe assessment. The Exudates (EXs) present in macula district are portioned proficiently by inclination based versatile thresholding. Deep Convolution neural network used for classification of DME and non-DME affected images. The division of OD and EXs has accomplished SE of 95.21%, 98.2%, SPE of 82.36%, 97.4% and ACC of 99.21%, 99.23% separately.

Diabetic Retinopathy (DR) is an adverse change in retinal blood vessels leads to blindness for diabetic patients without any symptoms. Diabetes is characterized by imbalance level of glucose in the human body. The optic disc (OD) is the major retinal landmark. Localization of OD is an important step in fundus image analysis and to develop Computer Aided Diagnosis tool for DR. OD center detection is necessary to reduce false positive rate in the detection of exudates (EXs). EXs is the white lesion present in the retina which is the early symptom for the diagnosis of DR. OD is detected using intensity variation algorithm and EXs is segmented using inverse surface adaptive thresholding algorithm. This algorithm achieves better result in localizing OD and segmenting EXs when compared to literature-reviewed methods. The maximum intensity variance method is used to locate OD with average ACC of 96.54%, 98.65%, 98.12%, 99.23%, 99.81% and 98.47% in DIARETDB0, DIARETDB1, MESSIDOR, DRIVE, STARE and Bejan Singh Eye Hospital databases with less computation time of 102 ms, 108 ms, 120 ms, 93 ms, 110 ms and 131 ms. The inverse surface adaptive thresholding method has achieved an SE of 97.43%, 98.87%, 99.12%, 97.21%, 98.72%, and 96.63%, a SPE of 91.56%, 92.31%, 90.21%, 90.14%, 89.58%, 92.56% and an ACC of 99.34%, 99.67%, 98.34%, 98.87%, 99.13%, 98.34% for DIARETDB0, DIARETDB1, MESSIDOR, DRIVE, STARE and Bejan Singh Eye Hospital databases respectively.

Cooperation among unlicensed secondary users was more significant in cognitive radio ad hoc networks (CRAHNs). The main challenges of CRAHNs now a day address the main issues of spectrum scarcity and underutilization. Due to the lack of channel sensing, cooperative spectrum sensing (CSS) is the main challenge in CRAHNs. Efficient CSS is more essential to enhance the detection of performance by utilizing the theoretical studies of spatially distributed secondary users (SUs). This paper uses the priority-based two-stage detection model (PBTSDM) to analyze the cooperation strategy among secondary users (SUs) and primary users (PUs). SUs in distributed CSS sense continuously amongst themselves and take unified decision on the presence or absence of PUs by using entropy-based energy detection mechanism. The simulation results reveal that the sensing accuracy in terms of sensing time and energy efficiency (in terms of minimum energy consumption) increased significantly by using the proposed scheme.

DAG scheduling is a process that plans and supervises the execution of interdependent tasks on heterogeneous computing resources. Efficient task scheduling is one of the important factors to improve the performance of heterogeneous computing systems. In this paper, an investigation on implementing Variable Neighborhood Search VNS algorithm for scheduling dependent jobs on heterogeneous computing and grid environments is carried out. Hybrid Two PHase VNS HTPHVNS DAG scheduling algorithm has been proposed. The performance of the VNS and HTPHVNS algorithm has been evaluated with Genetic Algorithm and Heterogeneous Earliest Finish Time algorithm. Simulation results show that VNS and HTPHVNS algorithm generally perform better than other meta-heuristics methods.

Cognitive Radio (CR) is a promising and potential technique to enable secondary users or unlicenced users to exploit the unused spectrum resources effectively possessed by primary users (PUs) or licenced users. The proven clustering approach is used to organize nodes in the network into logical groups to attain energy efficiency, network scalability, and stability for improving the sensing accuracy in CR through cooperative spectrum sensing. In this paper, a distributed dynamic load-balanced clustering algorithm is proposed. In this algorithm, each member in the cluster is to calculate the cooperative gain, residual energy, distance, and sensing cost from the neighboring clusters to perform the optimal decision. Each member in a cluster participates in selecting a cluster head through cooperative gain, and residual energy that minimise network energy consumption and enhances the channel sensing. First, we form the number of clusters using the Markov decision process model to reduce the energy consumption in a network. In this algorithm, CR users effectively utilize the PUs' reporting time slots when PUs are not available. The simulation results reveal that the cluster convergence, energy efficiency, and accuracy of channel sensing increased considerably by using the proposed algorithm.

Membrane proteins are vital type of proteins that serve as channels, receptors and energy transducers in a cell. They perform various important functions, which are mainly associated with their types. They are also attractive targets of drug discovery for various diseases. So predicting membrane protein types is a crucial and challenging research area in bioinformatics and proteomics. Because of vast investigation of uncharacterized protein sequences in databases, customary biophysical techniques are extremely tedious, costly and vulnerable to mistakes. Subsequently, it is very attractive to build a vigorous, solid, proficient technique to predict membrane protein types. In this work, a novel feature set Exchange Group Based Protein Sequence Representation (EGBPSR) is proposed for classification of membrane proteins with two new feature extraction strategies known as Exchange Group Local Pattern (EGLP) and Amino acid Interval Pattern (AIP). Imbalanced dataset and large dataset are often handled well by decision tree classifiers. Since imbalanced dataset are taken, the performance of various decision tree classifiers such as Decision Tree (DT), Classification and Regression Tree (CART), ensemble methods such as Adaboost, Random Under Sampling (RUS) boost, Rotation forest and Random forest are analyzed. The overall accuracy achieved in predicting membrane protein types is 96.45%.

This paper presents a new approach to segment pre-retinal and sub-retinal haemorrhage from colour fundus retinal image based on improved toboggan segmentation. Effective segmentation of pre-retinal, sub-retinal, small/blot haemorrhage detection is important in the development of screening system of Diabetic Retinopathy (DR). The strategy involves the application of edge detection and super-pixel based toboggan segmentation. The main advantage of the proposed work is that it does not depend on the manually selected parameters. This haemorrhages detection technique leads to high accuracy of 93.21% in comparison with the ground truth images from ophthalmologist.

Human Eye is one of the most sophisticated organ, with cornea ,retinal pigment epithelium, bruch membrane, macula, vitreous body, retina and optic nerve. Retinal image analysis is broadly used for screening the patients affected from sight threatening eye diseases like Diabetic Retinopathy (DR) and glaucoma. Optic disc (OD) margin is the mandatory landmark in establishing a frame of reference of classifying retinal and optic nerve pathology. Trustworthy and efficient OD localization and segmentation are important tasks in automatic eye disease screening. This paper presents a fully automated method for OD segmentation algorithm developed for retinal disease screening. First, preprocessing step used to remove noise by illumination correction and contrast enhancement. Second, a thresholding technique using Atanassov’s intuitionistic fuzzy set (A-IFS) is employed. This approach uses A-IFS histon, an encrustation of the histogram consist of the pixels that belong to the set of similar pixels, in roughness index A-IFS histon & histogram correlated to upper and lower approximations and optimum threshold value is identified which pixel belong to background or to object(OD). This approach is evaluated by means of three publicly available databases DRIVE, MESSIDOR, DIARETDB0, the experimental outcome shows that the overall performance is with 99% correct optic disc localization.

Glaucoma is an eye disease that leads to vision loss without any major symptoms. Detecting the disease in time is important because it cannot be cured. Due to raised Intra-Ocular Pressure (IOP), the Optic Nerve Head (ONH) gets damaged and the assessment of ONH is more promising because it consists of Optic Disc (OD) and Optic Cup (OC) in retinal fundus image. OD and OC detection is important preprocessing step to diagnose glaucoma. In this paper, we proposed a segmentation of OD by FeatureMatch (FM) technique and OD is localized using medial axis detection. The correspondence between the OD reference image and any input image is found by FM. This correspondence provides a distribution of patches in the input image that are closest to patches in the reference image. The proposed work is evaluated using DRIONS-DB that consists of 110 images. The OD is correctly localized in more than 60 images. The performance measures such as Root Mean Square Error (RMSE) and Mean of Structural SIMilarity (MSSIM) are calculated. The average computational time for this proposed work is about 14 secs. The average RMSE of the proposed work is about 0.05389.

Grid computing solves high-performance and high-throughput computing problems through sharing nodes ranging from personal computers to supercomputers distributed around the world. As the grid environments facilitate distributed computation, the scheduling of grid jobs has become an important issue. In this paper, an investigation on implementing Two-Phase Variable Neighborhood Search (TPVNS) algorithm for scheduling independent jobs on computational grid is carried out. The proposed algorithm consists of two modules with General Variable Neighborhood Search and Basic Variable Neighborhood Search algorithms in order to find a good mapping of grid jobs with grid nodes. The performance of the proposed algorithm has been evaluated with deterministic heuristic and evolutionary algorithms. Simulation results show that TPVNS algorithm generally performs better than the existing methods.