Multispectral image analysis for monitoring by IoT based wireless communication using secure locations protocol and classification by deep learning techniques

Present paper highlights the multispectral high-resolution satellite image has been classified using hybrid convolution neural network. Till now, improving the accuracy of image classification is one of the most significant research area in the domain of remote sensing. The most common deep learning technique such as convolution neural network (CNN) is used for image classification which becomes newer. Extraction of spatial and spectral information from satellite image using the 3D CNN approach is much more complex. While 2D CNN method is mainly used for extraction of spatial information, both spatial and spectral information are available in the multispectral satellite images. In this article, two CNN models (3D and 2D CNN) are integrated into hybrid CNN model which has been applied to multispectral satellite image for extraction of more precise land cover information. The classification results are validated using the overall accuracy and the Kappa statistic which was obtained through compared with the classified data and the Google Earth observation data. Hybrid CNN approach outcome is distinguished with the other methods such as fuzzy C-means (FCM), maximum likelihood classifier (MLC), and self-organizing maps (SOM). The classification accuracy for hybrid CNN model was found 95.17%, which is much higher than the other techniques.KeywordsMultispectral satellite imageClassificationHybrid CNNAccuracy

Machine learning methods are widely used today in the search for biologically active substances. Moreover, chemical and biological data have a highly specific structure, and all medicinal substances act simultaneously on several biotargets. Given this, the development of new methods for constructing deep-learning convolutional neural networks to analyze the relationships between multi-target biological activity and the structure of chemical compounds is a relevant and scientifically important task. Purpose of the work: To create a methodology for building multi-target convolutional neural networks of deep learning based on the correlation convolution of multiple docking energies into relevant biotargets. Materials and methods: Ensemble multiple docking of 537 compounds with anxiolytic activity and 234 compounds with antimicrobial activity against S.aureus into 22 and 10 relevant biotargets respectively, and the subsequent generation of their multiple docking energy spectra were performed using the original MSite program and AutoDock Vina program. Using the original FCCorNet program, correlation convolution of the energy spectra of multiple docking was performed and the energies of fully-connected convolutional neural networks were calculated for the specified compounds. The original computer DeepNets program for constructing deep-learning neural networks was developed in Python using the PyTorch library. Multi-target convolutional neural networks of deep learning were trained on two datasets, including the levels of anxiolytic activity and antimicrobial activity against S. aureus of known compounds and the energies of fully-connected convolutional correlation neural networks, and their accuracy was estimated. Results and discussion: The accuracy of the constructed neural network model for anxiolytic activity was Acc = 68.3 %, with a statistical significance of p = 1.1 × 10-9. The accuracy of the constructed neural network model for antimicrobial activity against S. aureus was Acc = 90.5 %, with a statistical significance of p 1 × 10-15. The accuracy of predicting antimicrobial activity for S. aureus exceeds that of predicting anxiolytic activity, possibly due to the more complex systemic multi-target mechanism underlying psychotropic effects, compared to the antibacterial action of chemical compounds. The results demonstrate the high validity of a new deep-learning convolutional neural network architecture for in silico searches for biologically active substances. Conclusions: A new multi-target deep-learning convolutional neural network architecture based on correlation convolution of energy spectra of multiple docking into a set of relevant biotargets has been developed. The developed methodology can be used for in silico searches for new high active compounds with various types of multi-target pharmacological activity.

Feasibility study on Huanglongbing (citrus greening) detection based on WorldView-2 satellite imagery

The use of machine learning models, particularly deep learning models, for the analysis of remote sensing products, especially multispectral satellite images, has recently experienced exponential development. Therefore, this article will present a protocol for processing multispectral satellite images by deep learning through the latest methods used in neural networks for computer vision, exploring all the methods used and proposed. In this study, we present the main methods of deep learning adapted to the processing of multispectral satellite images in the form of an efficient processing protocol. Our methodology proceeds with a systematic analysis of all the deep learning concepts by testing the applicability of multispectral satellite images and the contribution of the concept to the accuracy and performance of the model. In addition, each method introduced in this study has been tested in a real use case of remote sensing products especially satellite imagery for spatial analysis tasks such as semantic segmentation, object and pixel classification, object detection, image fusion, and land use and land cover classification (LULC). Thus, a discussion of the use of this protocol and some open challenges in this technological field are presented.

Edge enhancement in multispectral satellite images by means of vector operators

Recently, transforming windows files into images and its analysis using machine learning and deep learning have been considered as a state‐of‐the art works for malware detection and classification. This is mainly due to the fact that image‐based malware detection and classification is platform independent, and the recent surge of success of deep learning model performance in image classification. Literature survey shows that convolutional neural network (CNN) deep learning methods are successfully employed for image‐based windows malware classification. However, the malwares were embedded in a tiny portion in the overall image representation. Identifying and locating these affected tiny portions is important to achieve a good malware classification accuracy. In this work, a multi‐headed attention based approach is integrated to a CNN to locate and identify the tiny infected regions in the overall image. A detailed investigation and analysis of the proposed method was done on a malware image dataset. The performance of the proposed multi‐headed attention‐based CNN approach was compared with various non‐attention‐CNN‐based approaches on various data splits of training and testing malware image benchmark dataset. In all the data‐splits, the attention‐based CNN method outperformed non‐attention‐based CNN methods while ensuring computational efficiency. Most importantly, most of the methods show consistent performance on all the data splits of training and testing and that illuminates multi‐headed attention with CNN model's generalizability to perform on the diverse datasets. With less number of trainable parameters, the proposed method has achieved an accuracy of 99% to classify the 25 malware families and performed better than the existing non‐attention based methods. The proposed method can be applied on any operating system and it has the capability to detect packed malware, metamorphic malware, obfuscated malware, malware family variants, and polymorphic malware. In addition, the proposed method is malware file agnostic and avoids usual methods such as disassembly, de‐compiling, de‐obfuscation, or execution of the malware binary in a virtual environment in detecting malware and classifying malware into their malware family.

Land cover classification (LCC) using satellite images is one of the rapidly expanding fields in mapping, highlighting the need for updating existing computational classification methods. Advances in technology and the increasing variety of applications have introduced challenges, such as more complex classes and a demand for greater detail. In recent years, deep learning and Convolutional Neural Networks (CNNs) have significantly enhanced the segmentation of satellite images. Since the training of CNNs requires sophisticated and expensive hardware and significant time, using pre-trained networks has become widespread in the segmentation of satellite image. This study proposes a hybrid synergistic semantic segmentation method based on the Deeplab v3+ network and a clustering-based post-processing scheme. The proposed method accurately classifies various land cover (LC) types in multispectral satellite images, including Pastures, Other Built-Up Areas, Water Bodies, Urban Areas, Grasslands, Forest, Farmland, and Others. The post-processing scheme includes a spectral bag-of-words model and K-medoids clustering to refine the Deeplab v3+ outputs and correct possible errors. The simulation results indicate that combining the post-processing scheme with deep learning improves the Matthews correlation coefficient (MCC) by approximately 5.7% compared to the baseline method. Additionally, the proposed approach is robust to data imbalance cases and can dynamically update its codewords over different seasons. Finally, the proposed synergistic semantic segmentation method was compared with several state-of-the-art segmentation methods in satellite images of Italy’s Lake Garda (Lago di Garda) region. The results showed that the proposed method outperformed the best existing techniques by at least 6% in terms of MCC.

This paper presents a new deep learning based approach for soil detection using high resolution multispectral satellite images with a resolution of 0.31 m. In particular, a deep convolutional neural network (CNN) is proposed for soil detection to identify potential tunnel digging activities. Spatial and spectral information in the multispectral image cube has been incorporated into the CNN. We also propose a novel method to handle imbalance learning in the context of deep CNN model training. Experimental results on Worldview-2 (WV-2) multispectral satellite images captured at the border between USA and Mexico showed that the proposed CNN model can effectively detect soil in the remote sensed images, and the proposed imbalance learning technique improved the detection performance significantly.

The rapid evolution of telehealth, or telemedicine, has spurred crucial technological advancements aimed at addressing the early stages of complex cancer conditions, where conventional diagnostic methods face challenges. This research introduces a cancer detection system that utilizes Internet of Things (IoT)-based patient records and machine learning. The primary objective is to automate real-time breast cancer monitoring and detection in residential institutions and smart hospitals, thus enhancing the delivery of quality cancer healthcare. Background: Traditional diagnostic methods, particularly physical inspection, exhibit inherent limitations in identifying breast cancer at early stages. This research responds to this challenge by leveraging innovative technologies, such as IoT and deep learning-based techniques, to overcome the constraints of conventional approaches. Objective: The primary goal of this study is to develop and implement a cancer detection system that integrates IoT-based patient records and machine learning for real-time breast cancer monitoring in residential and healthcare settings. Method: The research employs a synergistic combination of IoT technology for collecting images of residential users and Convolutional Neural Network (CNN), a deep learning technique, for early cancer prediction. The focus lies on contributing to the overall well-being of individuals who may unknowingly be living with cancer. Result: Simulated outcomes after 25 epochs are presented, emphasizing the training accuracy of the model and its validation accuracy using the proposed VGG16 classifier. Graphical representations of the results indicate consistent performance metrics, with both validation and training accuracy exceeding 99%. Specifically, the training accuracy measures at an impressive 99.64%, while the validation accuracy stands at 99.12%. Main Findings: The study demonstrates the effectiveness of the integrated IoT and deep learning techniques in achieving high accuracy rates for early breast cancer prediction. The findings affirm the potential of this approach to assist dermatologists in identifying breast malignancies at treatable stages. Conclusion: This research establishes a foundational framework for the integration of IoT and deep learning techniques, presenting a promising avenue for advancing early cancer detection in smart healthcare systems. The proposed cancer detection system holds significant potential for improving healthcare outcomes and contributing to the overall well-being of individuals at risk of breast cancer.

The regular monitoring of agricultural areas is extremely important for mitigating food insecurity risks and for planning government interventions. In the literature, several deep learning algorithms have been recently proposed to perform land cover/ land use classification by using multispectral optical images. However, most of the considered deep learning models, such as the standard Convolutional Neural Networks (CNN), rely on mono-temporal images, focusing on spectral and textural features while discarding the temporal component, which is crucial for the accurate crop type mapping. In this work, we exploit a Long Short Term Memory (LSTM) deep learning classification architecture to characterize agricultural area dynamics by using the multitemporal multispectral information provided by satellite multispectral sensor Sentinel 2. Instead of considering a pre-trained network and applying to it a fine-tuning, the proposed architecture is trained from scratch in order to be tailored to the specific properties of the long time series of Sentinel 2 multispectral images. To face the lack of labeled training database, existing crop type maps available at the country level are used to generate a large set of weak reference data. First, the proposed method automatically extracts a large training dataset from existing crop type maps, by detecting those samples having the highest probability of being correctly classified. Then, the weak labeled samples extracted are used to train the deep LSTM architecture on a time series of Sentinel 2 images acquired over an entire year. The preliminary results obtained demonstrate the effectiveness of the proposed approach, which is promising at large scale.

Autonomous damage identification of submerged structure-foundation systems is challenging due to the difficulty of acquiring damage-induced system responses for training deep learning models. In this study, a novel approach integrating pseudo-damage simulation and convolutional neural network (CNN) deep learning is proposed for damage identification in the submerged structure-foundation system. Pseudo-damage simulation is a technique to generate equivalent damage conditions in inaccessible submerged sub-systems for training deep learning models. The following approaches are implemented to achieve the objective. Firstly, a scheme of pseudo-damage simulation for 1-D CNN deep learning is designed for the caisson-foundation system. Secondly, a vibration monitoring method using pseudo-wave-impulse excitations is designed for the caisson-foundation system. Thirdly, 1-D CNN models are trained for individual caisson units to predict the location and size of foundation damage by vibration signals out of a series of pseudo-damage cases. The 1-D CNN models demonstrate accurate performance in handling untrained scenarios. Experimental results validate the effectiveness of the proposed approach in achieving high accuracy for identification of the foundation damage.

The application of deep learning convolutional neural networks for solving the problem of automated facial expression recognition and determination of emotions of a person is analyzed. It is proposed to use the advantages of the transfer approach to deep learning convolutional neural networks training to solve the problem of insufficient data volume in sets of images with different facial expressions. Most of these datasets are labeled in accordance with a facial coding system based on the units ofhuman facial movement. The developed technology of transfer learning of the public deep learning convolutional neural networksfamilies DenseNet and MobileNet, with the subsequent “fine tuning”of the network parameters, allowed to reduce the training time and computational resources when solving the problem of facial expression recognition without losing the reliability of recognition of motor units. During the development of deep learning technology for convolutional neural networks, the following tasks were solved. Firstly, the choice of publicly available convolutional neural networks of the DenseNet and MobileNet families pre-trained on the ImageNet dataset was substantiated, taking into account the peculiarities of transfer learning for the task of recognizing facial expressions and determining emotions. Secondary, amodel of a deep convolutional neural network and a method for its training have been developed for solving problems of recognizing facial expressions and determining human emotions, taking into account the specifics of the selected pretrained convolutional neural networks. Thirdly, the developed deep learning technology was tested,and finally, the resource intensity and reliability of recognition of motor units on the DISFA set were assessed. The proposed technology of deep learning of convolutional neural networks can be used in the developmentof systems for automatic recognition of facial expressions and determination of humanemotions for both stationary and mobile devices. Further modification of the systems for recognizing motor units of human facial activity in order to increase the reliability of recognition is possible using of theaugmentation technique.

Land use recognition from multispectral satellite images is fundamentally critical for geological applications, but the results are not satisfied. The scale dimension of current multiscale learning is too coarse to account for rich scales in multispectral images, and pixel-wise classification tends to produce “salt-and-pepper” labels due to possible misclassification in heterogeneous regions. In this paper, these issues are addressed by proposing a new pixel-wise classification model with finer scales for convolutional neural networks. The model is designed to extract multiscale contextual information using multiscale networks at a fine-grained level, addressing the issue of insufficient multiscale learning for classification. Furthermore, a small-scale segmentation-combination method is introduced as a post-processing solution to smooth fragmented classification results. The proposed method is tested on GF-1, GF-2, DEIMOS-2, GeoEye-1, and Sentinel-2 satellite images, and compared with six neural-network-based algorithms. The results demonstrate the effectiveness of the proposed model in finding objects of large scale difference, improving classification accuracy, and reducing classified fragments. The discussion also illustrates that convolutional neural networks and pixel-wise inference are more practical than Transformer and patch-wise recognition.

Comprehensive Study for Breast Cancer Using Deep Learning and Traditional Machine Learning

PO-620-07 AN ENSEMBLE OF FEATURES BASED DEEP LEARNING NEURAL NETWORK FOR REDUCTION OF INAPPROPRIATE ATRIAL FIBRILLATION DETECTION IN IMPLANTABLE CARDIAC MONITORS