Comparison Of Convolutional Neural Networks Research Articles

Comparison of CNNs and Vision Transformers-Based Hybrid Models Using Gradient Profile Loss for Classification of Oil Spills in SAR Images

Oil spillage over a sea or ocean surface is a threat to marine and coastal ecosystems. Spaceborne synthetic aperture radar (SAR) data have been used efficiently for the detection of oil spills due to their operational capability in all-day all-weather conditions. The problem is often modeled as a semantic segmentation task. The images need to be segmented into multiple regions of interest such as sea surface, oil spill, lookalikes, ships, and land. Training of a classifier for this task is particularly challenging since there is an inherent class imbalance. In this work, we train a convolutional neural network (CNN) with multiple feature extractors for pixel-wise classification and introduce a new loss function, namely, “gradient profile” (GP) loss, which is in fact the constituent of the more generic spatial profile loss proposed for image translation problems. For the purpose of training, testing, and performance evaluation, we use a publicly available dataset with selected oil spill events verified by the European Maritime Safety Agency (EMSA). The results obtained show that the proposed CNN trained with a combination of GP, Jaccard, and focal loss functions can detect oil spills with an intersection over union (IoU) value of 63.95%. The IoU value for sea surface, lookalikes, ships, and land class is 96.00%, 60.87%, 74.61%, and 96.80%, respectively. The mean intersection over union (mIoU) value for all the classes is 78.45%, which accounts for a 13% improvement over the state of the art for this dataset. Moreover, we provide extensive ablation on different convolutional neural networks (CNNs) and vision transformers (ViTs)-based hybrid models to demonstrate the effectiveness of adding GP loss as an additional loss function for training. Results show that GP loss significantly improves the mIoU and F1 scores for CNNs as well as ViTs-based hybrid models. GP loss turns out to be a promising loss function in the context of deep learning with SAR images.

A comparison of convolutional neural networks for Kazakh sign language recognition

For people with disabilities, sign language is the most important means of communication. Therefore, more and more authors of various papers and scientists around the world are proposing solutions to use intelligent hand gesture recognition systems. Such a system is aimed not only for those who wish to understand a sign language, but also speak using gesture recognition software. In this paper, a new benchmark dataset for Kazakh fingerspelling, able to train deep neural networks, is introduced. The dataset contains more than 10122 gesture samples for 42 alphabets. The alphabet has its own peculiarities as some characters are shown in motion, which may influence sign recognition. Research and analysis of convolutional neural networks, comparison, testing, results and analysis of LeNet, AlexNet, ResNet and EffectiveNet – EfficientNetB7 methods are described in the paper. EffectiveNet architecture is state-of-the-art (SOTA) and is supposed to be a new one compared to other architectures under consideration. On this dataset, we showed that the LeNet and EffectiveNet networks outperform other competing algorithms. Moreover, EffectiveNet can achieve state-of-the-art performance on nother hand gesture datasets. The architecture and operation principle of these algorithms reflect the effectiveness of their application in sign language recognition. The evaluation of the CNN model score is conducted by using the accuracy and penalty matrix. During training epochs, LeNet and EffectiveNet showed better results: accuracy and loss function had similar and close trends. The results of EffectiveNet were explained by the tools of the SHapley Additive exPlanations (SHAP) framework. SHAP explored the model to detect complex relationships between features in the images. Focusing on the SHAP tool may help to further improve the accuracy of the model

Predicting sow postures from video images: Comparison of convolutional neural networks and segmentation combined with support vector machines under various training and testing setups

The use of CNN and segmentation to extract image features for the prediction of four postures for sows kept in crates was examined. The extracted features were used as input variables in an SVM classification method to estimate posture. The possibility of using a posture prediction model with images not necessarily obtained under the same conditions as those used for the training set was explored. As a reference case, the efficacy of the posture prediction model was explored when training and testing datasets were built using the same pool of images. In this case, all the models produced satisfactory results, with a maximum f1-score of 97.7% with CNNs and 93.3% with segmentation. To evaluate the impact of environmental variations, the models were trained and tested on different monitoring days. In this case, the best f1-score dropped to 86.7%. The impact of using the posture prediction model on animals that were not present in the training dataset was then explored. The best f1-score reduced to 63.4% when the posture prediction models were trained on one animal and tested on 11 other different animals. Conversely, when the models were tested on one animal and trained on the 11 others, the f1-score only decreased to 86% with the best model. On average, a decrease of around 17% caused by environmental and individual variations between training and testing was observed.

Comparison of Convolutional Neural Networks in Real-Time Monitoring of Aquaculture Water State

Real-time monitoring of the aquaculture water state is essential for shrimp aquaculture, and it plays a vital role in the growth of shrimp. The main purpose of this research is to find an effective method to achieve rapid detection and classification with water state images. We obtained a water state image dataset containing four categories in the laboratory, and gained the corresponding Hough datasets through Hough transform, including blue-green and turbid water, reddish brown and turbid water, light green and clear or tawny and clear water, colorless and transparent water. We compare the performance and spatio-temporal complexity of 12 widely used convolutional neural network models in water state image classification tasks, and then optimize the best performing model on the optimizer and hyperparameters. The experimental results show that the InceptionResNet model has the best effect and the InceptionV3 model has the smallest spatio-temporal complexity, while both models realize the task of water state image classification excellently.

Comparison of convolutional neural networks for detecting large vessel occlusion on computed tomography angiography.

Artificial intelligence diagnosis and triage of large vessel occlusion may quicken clinical response for a subset of time-sensitive acute ischemic stroke patients, improving outcomes. Differences in architectural elements within data-driven convolutional neural network (CNN) models impact performance. Foreknowledge of effective model architectural elements for domain-specific problems can narrow the search for candidate models and inform strategic model design and adaptation to optimize performance on available data. Here, we study CNN architectures with a range of learnable parameters and which span the inclusion of architectural elements, such as parallel processing branches and residual connections with varying methods of recombining residual information. We compare five CNNs: ResNet-50, DenseNet-121, EfficientNet-B0, PhiNet, and an Inception module-based network, on a computed tomography angiography large vessel occlusion detection task. The models were trained and preliminarily evaluated with 10-fold cross-validation on preprocessed scans (n=240). An ablation study was performed on PhiNet due to superior cross-validated test performance across accuracy, precision, recall, specificity, and F1 score. The final evaluation of all models was performed on a withheld external validation set (n=60) and these predictions were subsequently calibrated with sigmoid curves. Uncalibrated results on the withheld external validation set show that DenseNet-121 had the best average performance on accuracy, precision, recall, specificity, and F1 score. After calibration DenseNet-121 maintained superior performance on all metrics except recall. The number of learnable parameters in our five models and best-ablated PhiNet directly related to cross-validated test performance-the smaller the model the better. However, this pattern did not hold when looking at generalization on the withheld external validation set. DenseNet-121 generalized the best; we posit this was due to its heavy use of residual connections utilizing concatenation, which causes feature maps from earlier layers to be used deeper in the network, while aiding in gradient flow and regularization.

Deep Learning-Based Automated Lip-Reading: A Survey

A survey on automated lip-reading approaches is presented in this paper with the main focus being on deep learning related methodologies which have proven to be more fruitful for both feature extraction and classification. This survey also provides comparisons of all the different components that make up automated lip-reading systems including the audio-visual databases, feature extraction, classification networks and classification schemas. The main contributions and unique insights of this survey are: 1) A comparison of Convolutional Neural Networks with other neural network architectures for feature extraction; 2) A critical review on the advantages of Attention-Transformers and Temporal Convolutional Networks to Recurrent Neural Networks for classification; 3) A comparison of different classification schemas used for lip-reading including ASCII characters, phonemes and visemes, and 4) A review of the most up-to-date lip-reading systems up until early 2021.

Comparison of Convolutional Neural Networks for Dividing Seismic Sequences

Comparison of Convolutional Neural Networks for Dividing Seismic Sequences

Comparison of convolutional neural networks in fruit detection and counting: A comprehensive evaluation

Knowing beforehand the amount of fruit to be harvested leads to better logistics and decisions making in the agricultural industry. In the last years, several sensors –mainly artificial vision systems– and sensing techniques have been proposed to address the fruit counting problem with sundry results. Convolutional neural networks (CNN) arise as the current trend in processing imagery information, due to their adaptability and efficiency in object detection. However, there is still missing an insightful analysis of the usability of such technique in fruit counting problems in groves, since the learning process is sensitive to the training input data, the sensor (affected by environmental conditions) and the architecture chosen to process the imagery set. Therefore, in this work we test two of the most common architectures: Faster R-CNN with Inception V2 and Single Shot Multibox Detector (SSD) with MobileNet. These detection architectures were trained and tested on three fruits: Hass avocado and lemon (both from Chile), and apples (from California - USA), under different field conditions. To address the problem of video-based fruit counting, we use multi-object tracking based on Gaussian estimation. Our system achieves fruit counting performances up to 93% (overall for all fruits) using Faster-RCNN with Inception V2, and 90% (overall for all fruits) using SSD with MobileNet. Such results can lead to further improve the decision making process in agricultural practices.

Comparison of CNN and LSTM for Modeling Virtual Sensors in an Engine

<div class="section abstract"><div class="htmlview paragraph">The automotive industry makes extensive use of virtual models to increase efficiency during the development stage. The complexity of such virtual models increases with the complexity of the process that they describe, and thus new methods for their development are constantly evaluated. Among many others, data-driven techniques and machine learning offer promising results, creating deep neural networks that map complex input-output relations. This work aims at comparing the performance of two different neural network architectures for the estimation of the engine state and emissions (flow fuel, NOx and soot). More specifically, Convolutional Neural Network (CNN) and Long-Short Term Memory (LSTM) will be evaluated in terms of performance, using different techniques to increase the model generalization. During the learning stage data from different engine cycles are fed to the neural networks. In order to evaluate the generalization of the model, the networks are tested over new, previously unseen, engine cycles. Results show that our models over-perform other state-of-the-art models, the best performance was found for the LSTM model with 2.40%, 2.80% and 18.19% error for flow fuel, NOx and soot sensor respectively.</div></div>

Comparison of convolutional neural network and bag of features for multi-font digit recognition

<span>This paper evaluates the recognition performance of Convolutional Neural Network (CNN) and Bag of Features (BoF) for multiple font digit recognition. Font digit recognition is part of character recognition that is used to translate images from many document-input tasks such as handwritten, typewritten and printed text. BoF is a popular machine learning method while CNN is a popular deep learning method. Experiments were performed by applying BoF with Speeded-up Robust Feature (SURF) and Support Vector Machine (SVM) classifier and compared with CNN on Chars74K dataset. The recognition accuracy produced by BoF is just slightly lower than CNN where the accuracy of CNN is 0.96 while the accuracy of BoF is 0.94.</span>

Comparison of deep learning and human observer performance for detection and characterization of simulated lesions.

Detection and characterization of abnormalities in clinical imaging are of utmost importance for patient diagnosis and treatment. We present a comparison of convolutional neural network (CNN) and human observer performance on a simulated lesion detection and characterization task. We apply both conventional performance metrics, including accuracy and nonconventional metrics such as lift charts to perform qualitative and quantitative comparisons of each type of observer. It is determined that the CNN generally outperforms the human observers, particularly at high noise levels. However, high noise correlation reduces the relative performance of the CNN, and human observer performance is comparable to CNN under these conditions. These findings extend into the field of diagnostic radiology, where the adoption of deep learning is starting to become widespread. Consideration of the applications for which deep learning is most effective is of critical importance to this development.

Comparison of convolutional neural networks for landslide susceptibility mapping in Yanshan County, China

Assessments of landslide disasters are becoming increasingly urgent. The aim of this study is to investigate a convolutional neural network (CNN) framework for landslide susceptibility mapping (LSM) in Yanshan County, China. The two primary contributions of this study are summarized as follows. First, to the best of our knowledge, this report describes the first time that the CNN framework is used for LSM. Second, different data representation algorithms are developed to construct three novel CNN architectures. In this work, sixteen influencing factors associated with landslide occurrence were considered and historical landslide locations were randomly divided into training (70% of the total) and validation (30%) sets. Validation of these CNNs was performed using different commonly used measures in comparison to several of the most popular machine learning and deep learning methods. The experimental results demonstrated that the proportions of highly susceptible zones in all of the CNN landslide susceptibility maps are highly similar and lower than 30%, which indicates that these CNNs are more practical for landslide prevention and management than conventional methods. Furthermore, the proposed CNN framework achieved higher or comparable prediction accuracy. Specifically, the proposed CNNs were 3.94%–7.45% and 0.079–0.151 higher than those of the optimized support vector machine (SVM) in terms of overall accuracy (OA) and Matthews correlation coefficient (MCC), respectively.