Search

A deep learning network is super-efficient for image processing. The artificial brain of the deep learning system is highly capable of solving the categorization of problems. However, deep learning has several restrictions on the intelligence of the human brain. The human brain can update and improve knowledge by analyzing the features of images. In the artificial brain, the arrangements of the pixels are used for image understanding. This pixel-based continuous learning tendency of the artificial brain of deep learning leading to reducing the knowledge of previously achieved relevant information of the system is called the catastrophic forgetting problem of deep learning. Recently, many researchers reported that the catastrophic forgetting problem affects the result of deep learning. So, in this study, we are providing an excellent solution for the catastrophic forgetting problem of deep learning by restricting the direct analysis of input images with image morphological features. This image morphological feature-based deep learning system helps the artificial brain to concentrate on image features. This image morphological features system-based study achieved a stable learning graph with 80% average learning accuracy. [ABSTRACT FROM AUTHOR]

Presently, for the identification and classification of images, various deep learning techniques are being used. In these techniques, the whole image is considered to produce similar feature sets for many images. As a result, this mechanism loses many of its features at the final stage. Therefore, to analyze and identify medicinal leaves through an artificial eye of botanists, it was emphasized that the leaf image features should remain preserved till the final stage of classification for better accuracy. The existing plant identification approaches are trained using the leaf images. So leaf features are lost in the different stages of the convolution process and the same feature values are generated for similar type leaf images. This raises ambiguity in the results and affects the accuracy of leaf image identification. But here, in this proposed deep learning-based plant leaves morphological feature recognition system, leaf morphological features are used to train the system. Morphological features are identified to recognize a plant leaf. Here, morphological features of medicinal plant leaves, venation, shapes, apices, and bases are extracted and analyzed to predict the image class. So, the leaf features remain persevered until the final stage. The proposed feature recognition analysis improves the accuracy of the leaf identification method. In this, more than 300 leaves from 18 different plant families are collected and trained to build the deep learning classifier and achieve 96% accuracy. The performance evaluation was also conducted over "Flavia", "Swedish" and "Leaf" data set and obtained 91%, 87% and 91% accuracy. The performance of image classification and feature preservation algorithms with less computational power are indicating the potential applicability of the proposed Deep - Morpho Algorithm (DMA) in medicinal plants and leaves identification. [ABSTRACT FROM AUTHOR]

In the Covid-19 pandemic situation, the world is looking for immunity-boosting techniques for fighting against coronavirus. Every plant is medicine in one or another way, but Ayurveda explains the uses of plant-based medicines and immunity boosters for specific requirements of the human body. To help Ayurveda, botanists are trying to identify more species of medicinal immunity-boosting plants by evaluating the characteristics of the leaf. For a normal person, detecting immunity-boosting plants is a difficult task. Deep learning networks provide highly accurate results in image processing. In the medicinal plant analysis, many leaves are like each other. So, the direct analysis of leaf images using the deep learning network causes many issues for medicinal plant identification. Hence, keeping the requirement of a method at large to help all human beings, the proposed leaf shape descriptor with the deep learning-based mobile application is developed for the identification of immunity-boosting medicinal plants using a smartphone. SDAMPI algorithm explained numerical descriptor generation for closed shapes. This mobile application achieved 96%accuracy for the 64 × 64 sized images. [ABSTRACT FROM AUTHOR]

Among various types of skin diseases, skin cancer is the deadliest form of the disease. This paper classifies seven types of skin diseases: Actinic keratosis and intraepithelial carcinoma, Basal cell carcinoma, Benign keratosis, Dermatofibroma, Melanoma, Melanocytic type, and Vascular lesions. The primary objective of this paper is to evaluate the performance of these deep learning networks on skin lesion images. The lesion classification is implemented through transfer learning on fourteen deep learning networks: AlexNet, GoogleNet, ResNet50, VGG16, VGG19, ResNet101, InceptionV3, InceptionResNetV2, SqueezeNet, DenseNet201, ResNet18, MobileNetV2, ShuffleNet and NasNetMobile. The dataset used for these experiments are from ISIC 2018 of about 10,154 images. The results show that DenseNet201 performs best with 0.825 accuracy and improves skin lesion classification under multiple diseases. The proposed work shows the various parameters, including the accuracy of all fourteen deep learning networks, which helped build an efficient automated classification model for multiple skin lesions. [ABSTRACT FROM AUTHOR]

Over the past years, a digital multimedia uprising has been experienced in every walk of life, due to which the un-annotated or unstructured multimedia content has always been a key issue for research. The multimedia content is usually created with some intended emotions, which the creator wants to induce in viewers. The affectiveness of the multimedia content can be measured by analyzing elicited emotions of its viewers. In this paper, we present a rigorous study of human cognition using EEG signals while watching a video, to analyze the affectiveness of video content. The analysis presented in this paper is done to establish an effective relationship between video content and the human emotional state. For this, the most effective scalp location and frequency ranges are identified for two categories of videos, i.e., excited and sad. Furthermore, a common affective response (CAR) is extracted for finding the distinguishable features for aforementioned categories of videos. The CAR is calculated and tested on the publicly available dataset "AMIGOS," and the results presented here show the utility of cognitive features on extracted scalp locations and frequency ranges for automatic tagging of video content. The current research explores the innovative applicability of neuro-signals for a mouse-free video tagging based on human excitement level to augment a range of brain–computer interface (BCI)-based devices. It can further aid to automatically retrieve the video content which is exciting and interesting to human viewers. With this analysis, we aimed to provide a thorough analysis which can be used to customize a low-cost and mobile EEG system for automatic analysis and retrieval of videos. [ABSTRACT FROM AUTHOR]

The challenges of performance indices in highly complex computing have always been observed in terms of latency, throughput, number of gates, power, and yields. The use of CPU based vector processing and GPU parallel processing architecture has given ways to accelerate the computations but again lacks the scalability of parallel cores like Multiplier and Accumulator (MAC). With the evolution of reconfigurable hardware solutions and recent significant characterization in channel width (< 10 nm), power (zone-based partitions) and even cost of development, the FPGAs had come out of ages to be a most favorable choice not only for Proof of Concept (POC) or prototyping but as a product solution for many markets like medical, space and defense equipment. This work also proposes the FPGA based solution for hardware implementation of efficient computational techniques of de-convolution and image blending to cater the motion blurring problems in medical imaging. The active response time, taken during the real time image processing solutions, challenges the use case of many Image enhancement methods in the field. The proposed method and implementation scheme demonstrates the reduction on latency in motion de-blurring technique on hardware accelerated platform. [ABSTRACT FROM AUTHOR]

Numerous forms of skin diseases might attack the human body. The appropriate diagnoses and detection of the disease can decrease the severity level. The diseases of other categories are less dangerous when compared to skin cancer. Hence, an automated system is needed to classify skin disease from multiple categories with better accuracy. The paper proposes a classification model for various skin diseases using a four-approach system. To begin with, the use of transfer learning in pre-trained deep neural networks, namely, inceptionresnetv2, inceptionv3, resnet18, resnet101, vgg19, resnet50, and densenet201, are modified and trained for our system. The networks allowed extracting the features and apply using a multi-class support vector machine in the following approach. The former approach features are hybrid into two ensemble networks and applied using a multi-class support vector machine in the third approach. Finally, a shallow neural network is trained on the ensemble approach's features for classification in the fourth approach. The ISIC 2018 dataset is used for the experiments and evaluation of the proposed model. The results are analyzed with other states of the art classification methods, and our approach proves to be better with global accuracy at 96.2% and balanced accuracy at 89.2%. The receiver operating characteristic curve (ROC) shows the best performance of the proposed classification model. [ABSTRACT FROM AUTHOR]

Morphological features-based leaf identification algorithms provide highly accurate results. But it is required a single-lined edge extraction algorithm for morphological feature generation. Existing edge extraction algorithms have heavy calculations and higher iteration steps to extract edges. The simplicity of the edge detection algorithm helps to reduce the complexity of the image feature extraction process. In this paper, a fast and straightforward novel edge detection algorithm is introduced in the spatial domain. In a single iteration over all the pixels of the image, our algorithm can achieve a better result than existing edge detection techniques. Also, this paper provides a novel algorithm for leaf shape, vein, apex, and base feature extraction techniques using the edge detection algorithm that can be utilized further for the classification and identification of medicinal plant species or any other plant species too. The performance measure of the proposed edge detection algorithm for leaf features is better as compared to the existing edge detection algorithms. This edge detection algorithm achieved 92% of accuracy and a PSNR rate of 10.88 dB with the time complexity of O(n*m), where n is the height and m is the width of the given image. The importance of medicinal plant identification and existing leaf identification techniques are also discussed in this paper. [ABSTRACT FROM AUTHOR]

In hazy weather, the image in the scene suffers from noise which makes them less visible and to detect an object in hazy weather becomes a challenging task in computer vision. To have noise free image, many researchers have devised denoising techniques for enhancing visibility of images. Denoising is to remove the random variation from images and preserve the image features. As hazy images cause lots of visibility issues, this paper proposes removing haze and enhancing visibility of bad weather images with improved efficacy using an unsupervised neural network autoencoder that compress the data using machine learning and learns through Convolutional Neural Network (CNN). It has been observed that to have increased accuracy, the image classification and analysis is most effective using CNN. An end-to-end decoder training model is used to achieve the quality images. Further, various optimizers are compared to have better accuracy. The quality of images identified by estimation of performance such as RMSE and PSNR values are evaluated over single image and images from existing datasets and our own dataset. In the proposed method, RMSE value comes out to be 0.0373 for image from BSD500 dataset for specific image compared with other state of art approaches. The proposed model is intended in addition to other active, or progressive methods and the suggested method exceeds. The performance quality of images is explored applying measurable metrics. The images are taken from the datasets O-Haze, I-Haze, BSDS500, RESIDE, FRIDA and some from google. [ABSTRACT FROM AUTHOR]

Agriculture industry is the foundation of Indian economy where quality fruit production plays an important role. Apple or pome fruits are always in demand because of rich nutrients in it. Hence, to analyze and recognize the nutrients deficiency in fruits, a deep neural-based model is being proposed. This model automatically classifies and recognizes the type of deficiency present in apple. In this paper, a database has been created for four major types of nutrients deficiency in apples and used for training and validation of the proposed deep convolutional network. The model is tuned with k-fold cross-validation. The hyper-parameters such as epoch are set at 100 and batch size kept at 5. Finally, the model is tested with the testing data and achieved an average accuracy of 98.24% with k-fold cross-validation set to 15. The model accuracy depends on the hyper-parameters. The process of features optimization reduces the risk of overfitting of the model. Hence, careful selection of hyper-parameters is important for the convergence of cost function to the global minima that results in minimum misclassification. [ABSTRACT FROM AUTHOR]

Skin cancer is one of the diseases which lead to death if not detected at an early stage. Computer-aided detection and diagnosis systems are designed for its early diagnosis which may prevent biopsy and use of dermoscopic tools. Numerous researches have considered this problem and achieved good results. In automatic diagnosis of skin cancer through computer-aided system, feature extraction and reduction plays an important role. The purpose of this research is to develop computer-aided detection and diagnosis systems for classifying a lesion into cancer or non-cancer owing to the usage of precise feature extraction technique. This paper proposed the fusion of bag-of-feature method with speeded up robust features for feature extraction and quadratic support vector machine for classification. The proposed method shows the accuracy of 85.7%, sensitivity of 100%, specificity of 60% and training time of 0.8507 s in classifying the lesion. The result and analysis of experiments are done on the PH2 dataset of skin cancer. Our method improves performance accuracy with an increase of 3% than other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

It is a known fact that infrared radiation is produced by all objects with a temperature above absolute zero. In fruits, IR light sensor senses the invisible areas and exposes obscure objects in the image. In normal RGB images, it is very difficult to predict the internal defect of the fruit accurately without chopping it into pieces. In this paper, a thermal IR imaging-based fruit defect detection technique is proposed to identify and estimate the internal defect in the pome fruits. The technique is non-invasive and non-destructive which helps in minimizing the fruit wastage during the quality check. To achieve high accuracy, a three-level validation process is adopted. First level involves the prediction made by the proposed Deep learning-based expert system using RGB and thermal images of apple respectively. In second level, the validation of results is done using Fourier Transform Infrared spectroscopy technique. And finally, in third level, an invasive destructive method is used for inspection of the fruit quality by cutting them into pieces. The apple defect detection accuracy by the proposed Naïve Bayes classifier is observed to be 97.6% for thermal IR imaging samples whereas true color based achieved only 59% for the same sample. An apple seems to be healthy externally but internally there is a probability of a defect. Thermal IR imaging detects the heat from the surface of the fruit. Due to defective tissues of fruit pulp, the non-uniform temperature difference is observed and sensed on the surface of the fruit. [ABSTRACT FROM AUTHOR]

Background For ultrasonic signal generation and detection, the Capacitive Micromachined Ultrasonic Transducers (CMUTs), constructed using semiconductor fabrication processes are basically Microelectromechanical Systems based structures. Utilizing the semiconductor technology, the CMUTs can provide improved bandwidth, high transduction efficiency, ease of batch production etc. in contrast to piezoelectric transducers. Objective The CMUTs can be constructed in various possible geometries like square, rectangular, circular, and hexagonal. It has been found that CMUT with circular geometry provides optimum performance but has the disadvantage of area wastage in array construction due to voids in between the devices. Hence, the hexagonal CMUT is modelled and analyzed with circular CMUT. Method The Kirchhoff's thin plate theory is employed for novel analytical model of hexagonal CMUT. Result The results comparison is found similar with the circular CMUT with 13.42% of more wafer area utilization. Conclusion The proposed analytical model provides a cheaper design tool for the researchers with highly reduced simulation time. [ABSTRACT FROM AUTHOR]

Time synchronization is an essential characteristic for the faithful functioning of wireless sensor networks. Hypothetically, it is stated that the two clocks are considered to be synchronized with each other if their clock offset and frequency components within the oscillator are running at the same rate. But, practically, there is some degree of inaccuracy in the manufacturing of these clocks; as a result, their frequency sources have slight variations. This paper proposes a unique mathematical model which is based on maximum probable theory maximum probable clock offset estimation, along with the propagation delay factor for finding the best probable estimate for clock offset. In addition, a novel two-way message passing scheme-based network communication model is presented here. Based on results, it has been predicted that estimating the clock offset will help in improving the efficiency of the existing time-synchronization protocols i.e. time-sync protocol for sensor networks (TPSNs). It was observed that the standard error deviation of modified TPSN with MPCEO was reduced from 2.324 to 0.064 ms. This will make the modified TPSN model more accurate, efficient, and reliable. [ABSTRACT FROM AUTHOR]

Presently, the fruit industry requires a fast and efficient method for classification and recognition of the quality of fruits in bulk processing. Fruit recognition based on computer vision is quite challenging as it is based on the intensity, size, contour, and texture features extraction from fruits along with their suitable classifier selection. In this paper, the pixels containing the defected regions are segmented and their features are extracted. Further, a support vector machine (SVM) classifier is used to identify the defects and recognizes the cause with its stage. During the process of classification, fruits are categorized into two groups, defected and no-defect. The sample image observed defected further classified into three categories as the first, second and final stage of fruit defect. The sample testing at an early stage helps one to further proceed with the production or halt based on the outcome of a computer vision-based recognition system. [ABSTRACT FROM AUTHOR]

Synchronization of clock within a wireless sensor network epitomizes crucial problems in the efficient and reliable operation of the sensors. This paper discusses a novel probability theory based clock offset estimator for various clock synchronization schemes of wireless sensor networks is proposed. The motivation is to utilize local clock timing for achieving the global clock synchronization. It presents a probabilistic model to estimate the most expected value of clock offset for sensor nodes. This model uses a statistical tools based on dispersion and central tendency. The proposed model was compared with the existing clock offset estimating models. It was observed that the proposed model gives better results with 1.008% accuracy, 0.065% precision and 99.8% efficiency. [ABSTRACT FROM AUTHOR]

Synchronizing time between the sensors of wireless sensor network has vital importance. It helps in maintaining a consistent and reliable frame of time across the network. Two clocks are stated to be synchronized when their frequency source runs with equal rate and their offsets are set identical. Basically, due to the manufacturing difference there is slight variation in their clock oscillator which affects the degree of frequency source and accuracy. Therefore this leads to the problem of synchronizing time between the sensor clocks. To attain time synchronization in a network typical contention-based message passing techniques are used. In this paper two-way message passing scheme is utilized. It proposes a statistical tool based on the maximum probability theory for selecting the reference clock offset for time synchronization protocols. It also proposes a subset selection algorithm to support the proposed statistical tool. The results obtained consist of the selection of most probable estimate for clock offset. The proposed algorithm utilizes the two-way message passing scheme for the exchange of timing messages within the network. The proposed algorithm is compared with the existing algorithms for estimation of clock offset. It was observed that the proposed works gives better results in terms of efficiency i.e. 99.8% efficient. [ABSTRACT FROM AUTHOR]

Nowadays, multimedia content, such as photographs and movies, is ingrained in every aspect of human lives and has become a vital component of their entertainment. Multimedia content, such as videos or movie clips, is typically created with the intent to evoke certain feelings or emotions in viewers. Thus, by examining the viewer’s cognitive state while watching such content, its affectiveness can be evaluated. Considering the emotional aspect of videos, in this paper, a deep learning-based paradigm for affective tagging of video clips is proposed, in which participants’ irrational EEG responses are used to examine how people perceive videos. The information behind different brain regions, frequency waves, and connections among them play an important role in understanding a human’s cognitive state. Thus, here a contribution is made toward the effective modeling of EEG signals through two different representations, i.e., spatial feature matrix and combined power spectral density maps. The proposed feature representations highlight the spatial features of EEG signals and are therefore used to train a convolution neural network model for implicit tagging of two categories of videos in the Arousal domain, i.e., “Low Arousal” and “High Arousal.” The arousal emotional space represents the excitement level of the viewer; thus, this domain is selected to analyze the viewer’s engagement while watching video clips. The proposed model is developed using the EEG data taken from publicly available datasets “AMIGOS” and “DREAMER.” The model is tested using two different approaches, i.e., single-subject classification and multi-subject classification, and an average accuracy of 90%-95% and 90%-93% is achieved, respectively. The simulations presented in this paper show the pioneering applicability of the proposed framework for the development of brain–computer interface (BCI) devices for affective tagging of videos. [ABSTRACT FROM AUTHOR]

A pear is a sweet fruit that is rich in dietary fiber, antioxidants, and plant compounds. The nutritional disorder in pears is either due to deficiency of nutrients or toxicity of nutrients. The techniques to identify the nutrients deficiencies include tissue testing, soil analysis, plant analysis, and visual deficiency symptoms. The effects of alfalfa greening, black end, and cork spot are minimised by correcting the calcium nutrition in the pear tree. In this paper, a two-class decision jungle model is proposed for the recognition of calcium deficiency in pears based on a non-invasive approach. The calcium deficiency in pears makes a bumpy fruit surface and leaves yellow color on the affected area than the rest of the skin results in the greyish corky lesion. The nutrient deficiency that results in serious disorders in pears not only influences the plant but also impacts the fruit quality. The introduction of artificial intelligence in the agriculture industry has helped farmers to produce healthier fruits. The artificial intelligence provides a real-time data for the classifier that results in increasing agricultural efficiencies, better crop yields and reduce fruit production costs by facilitating the routine and most complex tasks. The two-class decision jungle model achieves an accuracy of 98% with a database of 1000 samples. The other approaches, such as Boosted decision tree, Bayes point machine, Logistic regression, Neural Network, and SVM, have an accuracy of 92.20%, 84.3%, 72.5%, 82.4%, and 72.5%, respectively for the equivalent datasets. The highest accuracy is achieved with the proposed two class decision jungle that has non-linear decision boundaries and the performance is resilient in the presence of features that consist of noise. The number of calcium-deficient and healthy pears is 500 each. The geometrical features are extracted for the development of an artificial intelligence-based model for the classification of two classes like calcium deficient and healthy pear. The extracted features are split into training, validation, and testing. For training, validation, and testing, 80%, 10% and 10% samples are used respectively. The precision level is observed to be 0.974 and test accuracy is achieved as 98.7% and the overall accuracy 98% which are better than the existing 88.2% accuracy for pears using Support Vector Machine. [ABSTRACT FROM AUTHOR]