Pre-trained AlexNet Research Articles

Preoperative prediction of microvascular invasion risk in hepatocellular carcinoma with MRI: peritumoral versus tumor region

ObjectivesTo explore the predictive performance of tumor and multiple peritumoral regions on dynamic contrast-enhanced magnetic resonance imaging (MRI), to identify optimal regions of interest for developing a preoperative predictive model for the grade of microvascular invasion (MVI).MethodsA total of 147 patients who were surgically diagnosed with hepatocellular carcinoma, and had a maximum tumor diameter ≤ 5 cm were recruited and subsequently divided into a training set (n = 117) and a testing set (n = 30) based on the date of surgery. We utilized a pre-trained AlexNet to extract deep learning features from seven different regions of the maximum transverse cross-section of tumors in various MRI sequence images. Subsequently, an extreme gradient boosting (XGBoost) classifier was employed to construct the MVI grade prediction model, with evaluation based on the area under the curve (AUC).ResultsThe XGBoost classifier trained with data from the 20-mm peritumoral region showed superior AUC compared to the tumor region alone. AUC values consistently increased when utilizing data from 5-mm, 10-mm, and 20-mm peritumoral regions. Combining arterial and delayed-phase data yielded the highest predictive performance, with micro- and macro-average AUCs of 0.78 and 0.74, respectively. Integration of clinical data further improved AUCs values to 0.83 and 0.80.ConclusionCompared with those of the tumor region, the deep learning features of the peritumoral region provide more important information for predicting the grade of MVI. Combining the tumor region and the 20-mm peritumoral region resulted in a relatively ideal and accurate region within which the grade of MVI can be predicted.Clinical relevance statementThe 20-mm peritumoral region holds more significance than the tumor region in predicting MVI grade. Deep learning features can indirectly predict MVI by extracting information from the tumor region and directly capturing MVI information from the peritumoral region.Key PointsWe investigated tumor and different peritumoral regions, as well as their fusion.MVI predominantly occurs in the peritumoral region, a superior predictor compared to the tumor region.The peritumoral 20 mm region is reasonable for accurately predicting the three-grade MVI.Graphical

Network traffic prediction by learning time series as images

Network traffic prediction is crucial for cost-effective network management, resource allocation, and security in the emergent software-defined and zero-touch networks. Machine Learning (ML) models such as Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) have shown great promise in traffic prediction. This work proposes a hybrid ML model consisting of pre-trained AlexNet and ResNet models, the 2D Convolutional Neural Network (CNN), Vision Transformer (ViT), and LSTM/GRU models and their variants to predict network traffic. This work explores two approaches for converting the time series data to images by allowing more precise feature extraction and then performing traffic prediction on an image dataset, thus increasing accuracy. Given their proficiency in extracting image features we use three pre-trained models: the AlexNet, the ResNet combined with CNN, and the ViT. Then, we use those models as inputs to the LSTM, Bi-LSTM, and the GRU, Bi-GRU models for predicting network traffic data. The models are evaluated on the GÉANT and Abilene traffic networks and assessed based on the Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). Comparison of results with existing network traffic prediction techniques shows considerable improvements in all performance metrics revealed by the Wilcoxon test. Our best model results for the GÉANT and Abilene datasets are respectively; MSE of 0.00058 and 0.03078, RMSE of 0.02415 and 0.17544, and MAE of 0.00774 and 0.13384.

A two-parameter extended logistic chaotic map for modern image cryptosystems

In this paper, a novel Extended Logistic Chaotic Map (ELCM) with two control parameters is proposed. Overcoming the major drawback of the standard Logistic and other existing chaotic maps, the ELCM not only has infinite chaotic range as well as good ergodicity, but also has a simple structure just like the Logistic map, which greatly facilitates its practical implementation and becomes very suitable for today's real-time applications. Moreover, a new color image encryption scheme based on chaos concept, DNA (Deoxyribonucleic Acid) encoding and Convolutional Neural Network (CNN) is also presented. To perform a chaotic scrambling, the ELCM is however used at different stages of the encryption process. In addition, a pre-trained AlexNet CNN is used to generate a public key. After XORing with the secret key, the latter is used, on the one hand, to generate the initials values and the control parameters of the ELCM. On the other hand, it will be split into two keys in order to generate a random grayscale image, which will then be XORed with the three components (i.e., R, G and B) of the color plaintext image. Afterward, a permutation operation, DNA encoding, diffusion operation as well as bit reversion operation are then applied to the R, G and B components. The performance evaluation demonstrates that the ELCM not only has an infinite chaotic parameter range, but also exhibits a high chaotic complexity. Besides, experimental results as well as security analysis confirm that with NPCR of 99.6287%, the designed color image encryption scheme achieves an average entropy of 7.9975 and a near zero correlation (-0.0027). Furthermore, the proposed encryption scheme is in fact of higher security level in comparison to other schemes recently presented in the literature, thus making it highly suitable for today's real-time applications.

Collision avoidance of mobile robot using Alexnet and NVIDIA Jetson Nano B01

In this research, an intelligence collision avoidance system on a mobile robot was designed using the AlexNet image classifier method. AlexNet is a convolutional neural network architecture that managed to win the ImageNet Large Scale Visual Recognition Challenge in 2012. The dataset consists of three categorical labels: blocked right, blocked left, and free. Images of 224 x 224 pixels were trained into two CNN architectures: AlexNet and ResNet-18. The performance of both architectures was examined in a testing environment. The system was built without real-time obstacles, instead using the side boundaries of the test lane. Analogously, if the mobile robot moves either through the side lane or off track, then these conditions are defined as a crash. From the entire research that was done, it was determined that intelligence collision avoidance models based on AlexNet were the most reliable models, with an average accuracy deviation rate of 6,00%. The true pre-trained AlexNet adopted from PyTorch Transfer Learning had 92.22% overall accuracy, while the non-trained AlexNet achieved 90.81% accuracy. It is also supported by the evidence that Intelligence Collision Avoidance Model-1 and Model-3 based on AlexNet didn’t lead the mobile robot to spin out and were stable in the test lane.

Classifying the Percentage of Broken Magnets in Permanent Magnet Synchronous Motors Using Combined Short-Time Fourier Transform and a Pre-Trained Convolutional Neural Network

In critical applications of electrical machines, ensuring validity and safety is paramount to prevent system failures with potentially hazardous consequences. The integration of machine learning (ML) technologies plays a crucial role in monitoring system performance and averting failures. Among various motor types, permanent magnet synchronous motors (PMSMs) are widely favored for their versatile speed range, enhanced power density, and ease of control, finding applications in both industrial settings and electric vehicles. This study focuses on the detection and classification of the percentage of broken magnets in PMSMs using a pre-trained AlexNet convolutional neural network (CNN) model. The dataset was generated by combining finite element methods (FEMs) and short-time Fourier transform (STFT) applied to stator phase currents, which exhibited significant variations due to diverse broken magnet structures. Leveraging transfer learning, the pre-trained AlexNet model underwent adjustments, including the elimination and rearrangement of the final three layers and the introduction of new layers tailored for electrical machine applications. The resulting pre-trained CNN model achieved a remarkable performance, boasting a 99.94% training accuracy and 0.0004% training loss in the simulation dataset, utilizing a PMSM with 4% magnet damage for experimental validation. The model’s effectiveness was further affirmed by an impressive 99.95% area under the receiver operating characteristic (ROC) curve in the experimental dataset. These results underscore the efficacy and robustness of the proposed pre-trained CNN method in detecting and classifying the percentage of broken magnets, even with a limited dataset.

Disease Classification of CXR Images by AlexNet Transfer Learning

The purpose of this study is to classify COVID-19 CXR(Chest X-Ray) images using a CNN(Convolutional Neural Network) modified from AlexNet using a transfer learning method so that it can be used as an auxiliary method when diagnosing diseases in medical settings. The Dataset used in the experiment uses "COVIDGR-1.0 Dataset". "COVIDGR-1.0 Dataset" is a collection of 784 anonymized X-ray images and, in collaboration with a team of radiologists, built into the dataset in accordance with labeling protocols.The proposed CNN is a fine-tuned build of pre-trained AlexNet, which takes the "COVIDGR-1.0 dataset" as input and categorizes it into Normal, Mild, Moderate and Sever according to the labels and categories. Experimental results of the disease classification of CXR images show that the AUC(Area Under the Curve) is 0.8781, with accuracy of 87.81%. It is believed that the results of this study can be used in the following fields. First, it can be used as an auxiliary tool for diagnosing lung diseases. Second, if the data set is ready, it can be used to classify other diseases. Third, data can be stored separately through classification. Fourth, in the medical field, the object to be classified can be imaged using a smartphone and then classified and stored. This study has the following limitations. The problem with automatically classifying diseases is the vastness and integrity of the dataset. If the data is not prepared, it will cause problems with the accuracy of classification. And in the medical field, doctors' decision-making (diagnosis) is not limited to image data. There is a limitation in that doctors make decisions based on comprehensive data and that medical decisions cannot be made by machines without the doctor's intervention. Determining the decision threshold in decision making problems by AI is also an important limitation. This is because setting reference points for sensitivity and specificity is not the domain of AI experts but of experts in the field. Therefore, medical staff in the field must participate in the process of developing AI in the medical field. Future tasks include continuing research in connection with program input stage development, performance improvement, app development, and medical big data servers.

USAGE OF ARTIFICIAL NEURAL NETWORKS IN THE DIAGNOSIS OF KNEE JOINT DISORDERS

Following article address the issue of automatic knee disorder diagnose with usage of neural networks. We proposed several hybrid neural net architectures which aim to successfully classify abnormality using MRI (magnetic resonance imaging) images acquired from publicly available dataset. To construct such combinations of models we used pretrained Alexnet, Resnet18 and Resnet34 downloaded from Torchvision. Experiments showed that for certain abnormalities our models can achieve up to 90% accuracy.

Hybrid Arabic handwritten character segmentation using CNN and graph theory algorithm

Arabic handwritten words segmentation is considered a challenging step in Arabic optical character recognition. In this work, we propose a convolutional neural network-based segmentation of handwritten Arabic words. After generating candidate segmentation points, we applied a graph theory-based technique to divide the word into sub-words. Our approach reached a 96% correct sub-word segmentation rate. Additionally, we develop a Segmentation Hypothesis Graph (SHG) to generate candidate characters for each sub-word. We employ both a manually designed convolutional neural network and a transfer learning-based CNN using the pretrained AlexNet model. The accuracy, 96.97%, of the pretrained model outperformed the manually and state of the art models trained on the same dataset. Finally, we used recognition confidence to validate the segmentation points. Hence, our approach correctly segments 88% of the words into characters and successfully resolves 100% of the overlapping cases.

AlexNet-Based Feature Extraction for Cassava Classification: A Machine Learning Approach

Cassava, a significant crop in Africa, Asia, and South America, is a staple food for millions. However, classifying cassava species using conventional color, texture, and shape features is inefficient, as cassava leaves exhibit similarities across different types, including toxic and non-toxic varieties. This research aims to overcome the limitations of traditional classification methods by employing deep learning techniques with pre-trained AlexNet as the feature extractor to accurately classify four types of cassava: Gajah, Manggu, Kapok, and Beracun. The dataset was collected from local farms in Lamongan Indonesia. To collect images with agricultural research experts, the dataset consists of 1,400 images, and each type of cassava has 350 images. Three fully connected (FC) layers were utilized for feature extraction, namely fc6, fc7, and fc8. The classifiers employed were support vector machine (SVM), k-nearest neighbors (KNN), and Naive Bayes. The study demonstrated that the most effective feature extraction layer was fc6, achieving an accuracy of 90.7% with SVM. SVM outperformed KNN and Naive Bayes, exhibiting an accuracy of 90.7%, sensitivity of 83.5%, specificity of 93.7%, and F1-score of 83.5%. This research successfully addressed the challenges in classifying cassava species by leveraging deep learning and machine learning methods, specifically with SVM and the fc6 layer of AlexNet. The proposed approach holds promise for enhancing plant classification techniques, benefiting researchers, farmers, and environmentalists in plant species identification, ecosystem monitoring, and agricultural management.

Unleash the Power of Deep Neural Networks and Transfer Learning for Enhanced Face Recognition

Face recognition has become a crucial technology with diverse applications in various domains such as biometric security, healthcare, personal identification, and law enforcement. It stands out due to its high recognition rate, the richness of features, its universality, and its uniqueness. Having said that, challenges such as varying illumination, poses, facial expressions, face transformations, and backgrounds have hindered recognition accuracy, especially when using traditional techniques that rely on explicit feature extraction before the actual classification. The emergence of deep learning algorithms, such as Convolutional Neural Networks (CNNs), has exhibited prominent potential in image recognition capabilities, but building effective face recognition systems from scratch demands substantial labeled data and computational resources. To overcome these challenges, this paper proposes a novel approach that utilizes transfer learning and a pre-trained AlexNet CNN to improve face recognition capabilities. The model is trained with three different deep-learning optimizers, and the most suitable one is selected for the proposed system. For experiments, we created a custom database with images of five distinct individuals, generating over 1260 face images with each person having at least 250 samples. The dataset was further enriched using data augmentation technique. Using MATLAB with the Deep Learning Toolbox, the results demonstrate the model's recognition rate, peaking up to 100 per cent when the SGD optimizer is employed, 97.88 per cent when RMSProp is the optimizer, and 99.2 per cent for the Adam optimizer while using the same set of hyperparameters. These findings showcase the effectiveness of the approach and open doors for more efficient and accurate face recognition systems in real-world scenarios with large datasets.

Deep Learning-based Copy-Move and Spliced Image Forgery Detection

This paper proposes a Deep Learning (DL) based pre-trained AlexNet model for detecting and localizing copy-move and spliced forgery in photos. To localise forgeries in a photo, a binary mask is constructed using sobel operators. Further feature vectors are extracted patch wise from the input pictures. The Spatial Rich Model (SRM) is employed to address the generalisation issues in the DL model. There are three datasets used: Columbia Uncompressed Image Splicing Detection Evaluation (CUISDE), CASIA 1, and CASIA 2. The overall performance of the model has a 98.59 percent total accuracy as against 98.176% reported in the existing literature.

Computer-aided Detection of Tuberculosis from Microbiological and Radiographic Images

ABSTRACT Tuberculosis caused by Mycobacterium tuberculosis have been a major challenge for medical and healthcare sectors in many underdeveloped countries with limited diagnosis tools. Tuberculosis can be detected from microscopic slides and chest X-ray but as a result of the high cases of tuberculosis, this method can be tedious for both Microbiologists and Radiologists and can lead to miss-diagnosis. These challenges can be solved by employing Computer-Aided Detection (CAD)via AI-driven models which learn features based on convolution and result in an output with high accuracy. In this paper, we described automated discrimination of X-ray and microscope slide images into tuberculosis and non-tuberculosis cases using pretrained AlexNet Models. The study employed Chest X-ray dataset made available on Kaggle repository and microscopic slide images from both Near East University Hospital and Kaggle repository. For classification of tuberculosis using microscopic slide images, the model achieved 90.56% accuracy, 97.78% sensitivity and 83.33% specificity for 70: 30 splits. For classification of tuberculosis using X-ray images, the model achieved 93.89% accuracy, 96.67% sensitivity and 91.11% specificity for 70:30 splits. Our result is in line with the notion that CNN models can be used for classifying medical images with higher accuracy and precision.

Deep-Ocular: Improved Transfer Learning Architecture Using Self-Attention and Dense Layers for Recognition of Ocular Diseases.

It is difficult for clinicians or less-experienced ophthalmologists to detect early eye-related diseases. By hand, eye disease diagnosis is labor-intensive, prone to mistakes, and challenging because of the variety of ocular diseases such as glaucoma (GA), diabetic retinopathy (DR), cataract (CT), and normal eye-related diseases (NL). An automated ocular disease detection system with computer-aided diagnosis (CAD) tools is required to recognize eye-related diseases. Nowadays, deep learning (DL) algorithms enhance the classification results of retinograph images. To address these issues, we developed an intelligent detection system based on retinal fundus images. To create this system, we used ODIR and RFMiD datasets, which included various retinographics of distinct classes of the fundus, using cutting-edge image classification algorithms like ensemble-based transfer learning. In this paper, we suggest a three-step hybrid ensemble model that combines a classifier, a feature extractor, and a feature selector. The original image features are first extracted using a pre-trained AlexNet model with an enhanced structure. The improved AlexNet (iAlexNet) architecture with attention and dense layers offers enhanced feature extraction, task adaptability, interpretability, and potential accuracy benefits compared to other transfer learning architectures, making it particularly suited for tasks like retinograph classification. The extracted features are then selected using the ReliefF method, and then the most crucial elements are chosen to minimize the feature dimension. Finally, an XgBoost classifier offers classification outcomes based on the desired features. These classifications represent different ocular illnesses. We utilized data augmentation techniques to control class imbalance issues. The deep-ocular model, based mainly on the AlexNet-ReliefF-XgBoost model, achieves an accuracy of 95.13%. The results indicate the proposed ensemble model can assist dermatologists in making early decisions for the diagnosing and screening of eye-related diseases.

Characterization of ink-based phantoms with deep networks and photoacoustic method

<p class="abstract">This study aims to explore the feasibility of using an in-house developed photoacoustic (PA) system for predicting blood phantom concentrations using a pretrained Alexnet and a Long Short-Term Memory (LSTM) network. In two separate experiments, we investigate the performance of our strategy using a point laser source and a color-tunable Light-Emitting Diode (LED) as the illumination source. A single-point transducer is employed to measure signal change by adding ten different black ink concentrations into a tube. These PA signals are used for training and testing the employed deep networks. We found that the LED system with light wavelength of 450 nm gives the best characterization performance. The classification accuracy of the Alexnet and LSTM models tested on this dataset shows an average value of 94% and 96%, respectively, making this a preferred light wavelength for future operation. Our system may be used for the noninvasive assessment of microcirculatory changes in humans.</p>

Teeth Lesion Detection Using Deep Learning and the Internet of Things Post-COVID-19.

With a view of the post-COVID-19 world and probable future pandemics, this paper presents an Internet of Things (IoT)-based automated healthcare diagnosis model that employs a mixed approach using data augmentation, transfer learning, and deep learning techniques and does not require physical interaction between the patient and physician. Through a user-friendly graphic user interface and availability of suitable computing power on smart devices, the embedded artificial intelligence allows the proposed model to be effectively used by a layperson without the need for a dental expert by indicating any issues with the teeth and subsequent treatment options. The proposed method involves multiple processes, including data acquisition using IoT devices, data preprocessing, deep learning-based feature extraction, and classification through an unsupervised neural network. The dataset contains multiple periapical X-rays of five different types of lesions obtained through an IoT device mounted within the mouth guard. A pretrained AlexNet, a fast GPU implementation of a convolutional neural network (CNN), is fine-tuned using data augmentation and transfer learning and employed to extract the suitable feature set. The data augmentation avoids overtraining, whereas accuracy is improved by transfer learning. Later, support vector machine (SVM) and the K-nearest neighbors (KNN) classifiers are trained for lesion classification. It was found that the proposed automated model based on the AlexNet extraction mechanism followed by the SVM classifier achieved an accuracy of 98%, showing the effectiveness of the presented approach.

Classification of Appearance Quality of Red Grape Based on Transfer Learning of Convolution Neural Network

Grapes are a globally popular fruit, with grape cultivation worldwide being second only to citrus. This article focuses on the low efficiency and accuracy of traditional manual grading of red grape external appearance and proposes a small-sample red grape external appearance grading model based on transfer learning with convolutional neural networks (CNNs). In the naturally growing vineyards, 195,120,135 samples of Grade I, Grade II, and Grade III red grapes were collected using a Canon EOS 550D camera, and a data set of 1800 samples was obtained using data enhancement technology. Then, the CNN transfer learning method was used to transfer the pre-trained AlexNet, VGG16, GoogleNet, InceptionV3, and ResNet50 network models on the ImageNet image dataset to the red grape image grading task. By comparing the classification performance of the CNN models of these five different network depths with fine-tuning, ResNet50 with a learning rate of 0.001 and a loop number of 10 was determined to be the best feature extractor for red grape images. Moreover, given the small number of red grape image samples in this study, different convolutional layer features output by the ResNet50 feature extractor were analyzed layer by layer to determine the effect of deep features extracted by each convolutional layer on Support Vector Machine (SVM) classification performance. This analysis helped to obtain a ResNet50 + SVM red grape external appearance grading model based on the optimal ResNet50 feature extraction strategy. Experimental data showed that the classification model constructed using the feature parameters extracted from the 10th node of the ResNet50 network achieved an accuracy rate of 95.08% for red grape grading. These research results provide a reference for the online grading of red grape clusters based on external appearance quality and have certain guiding significance for the quality and efficiency of grape industry circulation and production.

Alexnet architecture variations with transfer learning for classification of wound images

In medical world, wound care and follow-up is one of the issues that are gaining importance to work on day by day. Accurate and early recognition of wounds can reduce treatment costs. In the field of computer vision, deep learning architectures have received great attention recently. The achievements of existing pre-trained architectures for describing (classifying) data belonging to many image sets in the real world are primarily addressed. However, to increase the success of these architectures in a certain area, some improvements and enhancements can be made on the architecture. In this paper, the classification of pressure and diabetic wound images was performed with high accuracy. The six different new AlexNet architecture variations (3Conv_Softmax, 3Conv_SVM, 4Conv_Softmax, 4Conv_SVM, 6Conv_Softmax, 6Conv_SVM) were created with a different number of implementations of Convolution, Pooling, and Rectified Linear Activation (ReLU) layers. Classification performances of the proposed models are investigated by using Softmax classifier and SVM classifier separately. A new original Wound Image Database are created for performance measures. According to the experimental results obtained for the Database, the model with 6 Convolution layers (6Conv_SVM) was the most successful method among the proposed methods with 98.85% accuracy, 98.86% sensitivity, and 99.42% specificity. The 6Conv_SVM model was also tested on diabetic and pressure wound images in the public medetec dataset, and 95.33% accuracy, 95.33% sensitivity, and 97.66% specificity values were obtained. The proposed method provides high performance compared to the pre-trained AlexNet architecture and other state-of-the-art models in the literature. The results showed that the proposed 6Conv_SVM architecture can be used by the relevant departments in the medical world with good performance in medical tasks such as examining and classifying wound images and following up the wound process.

Recognition of human action for scene understanding using world cup optimization and transfer learning approach.

Understanding human activities is one of the vital steps in visual scene recognition. Human daily activities include diverse scenes with multiple objects having complex interrelationships with each other. Representation of human activities finds application in areas such as surveillance, health care systems, entertainment, automated patient monitoring systems, and so on. Our work focuses on classifying scenes into different classes of human activities like waving hands, gardening, walking, running, etc. The dataset classes were pre-processed using the fuzzy color stacking technique. We adopted the transfer learning concept of pretrained deep CNN models. Our proposed methodology employs pretrained AlexNet, SqueezeNet, ResNet, and DenseNet for feature extraction. The adaptive World Cup Optimization (WCO) algorithm is used halfway to select the superior dominant features. Then, these dominant features are classified by the fully connected classifier layer of DenseNet 201. Evaluation of the performance matrices showed an accuracy of 94.7% with DenseNet as the feature extractor and WCO for feature selection compared to other models. Also, our proposed methodology proved to be superior to its counterpart without feature selection. Thus, we could improve the quality of the classification model by providing double filtering using the WCO feature selection process.

Ensuring privacy protection in the era of big laparoscopic video data: development and validation of an inside outside discrimination algorithm (IODA)

BackgroundLaparoscopic videos are increasingly being used for surgical artificial intelligence (AI) and big data analysis. The purpose of this study was to ensure data privacy in video recordings of laparoscopic surgery by censoring extraabdominal parts. An inside-outside-discrimination algorithm (IODA) was developed to ensure privacy protection while maximizing the remaining video data.MethodsIODAs neural network architecture was based on a pretrained AlexNet augmented with a long-short-term-memory. The data set for algorithm training and testing contained a total of 100 laparoscopic surgery videos of 23 different operations with a total video length of 207 h (124 min ± 100 min per video) resulting in 18,507,217 frames (185,965 ± 149,718 frames per video). Each video frame was tagged either as abdominal cavity, trocar, operation site, outside for cleaning, or translucent trocar. For algorithm testing, a stratified fivefold cross-validation was used.ResultsThe distribution of annotated classes were abdominal cavity 81.39%, trocar 1.39%, outside operation site 16.07%, outside for cleaning 1.08%, and translucent trocar 0.07%. Algorithm training on binary or all five classes showed similar excellent results for classifying outside frames with a mean F1-score of 0.96 ± 0.01 and 0.97 ± 0.01, sensitivity of 0.97 ± 0.02 and 0.0.97 ± 0.01, and a false positive rate of 0.99 ± 0.01 and 0.99 ± 0.01, respectively.ConclusionIODA is able to discriminate between inside and outside with a high certainty. In particular, only a few outside frames are misclassified as inside and therefore at risk for privacy breach. The anonymized videos can be used for multi-centric development of surgical AI, quality management or educational purposes. In contrast to expensive commercial solutions, IODA is made open source and can be improved by the scientific community.

Real-time monitoring of partial shading in large PV plants using Convolutional Neural Network

The performance of a typical string-connected Photovoltaic (PV) panels will be adversely affected even if only one of its panels is partially shaded. Therefore, the monitoring of these panels becomes crucial to ensure reliable operation of the whole PV plant. In this paper, a low-cost solution for real-time monitoring and diagnosis of PV plants is proposed. In fact, our approach is based on the use of pre-trained AlexNet Convolutional Neural Network (CNN) to predict whether a given panel is under partial shading conditions (PSC) or not, by extracting features from two-dimensional (2-D) scalograms generated in real-time from PV data acquisition system. It should be noted that, in addition to a disconnection circuit and a servo motor mounted on it, each panel of the proposed system is equipped with four sensors namely: voltage, current, temperature and irradiance sensor. All these components are connected to Adafruit PyBadge microcontroller unit (MCU). In fact, a Continuous Wavelet Transform was applied to the time series data provided by these sensors to generate the 2-D scalograms. On the hand, we performed transfer learning by fine-tuning the last fully connected layer of the pre-trained AlexNet CNN and modifying its architecture to get only two neurons which represent our two classes (i.e., PV panel under PSC or not). The model, achieving a high fault detection accuracy of 98.05%, was built in Keras framework with TensorFlow where only the parameters of the last fully connected layer of the pre-trained AlexNet CNN are retrained. Afterward, this model was converted into TensorFlow Lite format and successfully implemented on PyBadge MCU. Several experiments were carried out in this work have shown that, our monitoring system allows not only the automatic disconnection of the partial shaded PV panels, to avoid thermal issues under bypass conditions, but also removing undesired objects accumulated on their surface by actioning the corresponding servo motor, in order to restore in real-time the normal operation of the whole PV system.