Remote Sensing Data Classification Using A Hybrid Pre-Trained VGG16 CNN- SVM Classifier

Tun, Nyan Linn; Gavrilov, A. I.; Tun, Naing Min; Trieu, Do Minh; Aung, Htet

doi:10.1109/elconrus51938.2021.9396706

Cited by 19 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This also keeps the ResNet model from overfitting because it is possible for the model to remember the features in the input. VGG16 is a model built using the overfit dataset consisting of 1.2 million images with 1000 classes used during model training [30], which consists of 13 convolutional layers, 3 fully connected layers for classification [43].…”

Section: Methodsmentioning

confidence: 99%

A Comparative Study of Javanese Script Classification with GoogleNet, DenseNet, ResNet, VGG16 and VGG19

Susanto,

Sari,

Rachmawanto

et al. 2024

SJI

View full text Add to dashboard Cite

Purpose: Javanese script is a legacy of heritage or heritage in Indonesia originating from the island of Java needs to be preserved. Therefore, in this study, the classification and identification process of Javanese script letters will be carried out using the CNN method. The purpose of this research is to be able to build a model which can properly classify Javanese script, it can help in the process of recognizing letters in Javanese script easily.Methods: In this study, the Javanese script classification process has been used the transfer learning process of Convolutional Neural Network, namely GoogleNet, DenseNet, ResNet, VGG16 and VGG19. The purpose of using transfer learning is to improve the sequential CNN model, processing can be better and optimal because it utilizes a previously trained model.Result: The results obtained after testing in this study are using the transfer learning method, the GoogleNet model gets an accuracy of 88.75%, the DenseNet model gets an accuracy of 92%, the ResNet model gets an accuracy of 82.75%, the VGG16 model gets an accuracy of 99.25% and the VGG19 model gets an accuracy of 99.50%.Novelty: In previous studies, it is still very rare to discuss the Javanese script classification process using the CNN transfer learning method and which method is the most optimal for performing the Javanese script classification process. In this study, it had been resulted find an effective method to be able to carry out the Javanese script classification process properly and optimally.

show abstract

Section: Methodsmentioning

confidence: 99%

A Comparative Study of Javanese Script Classification with GoogleNet, DenseNet, ResNet, VGG16 and VGG19

Susanto,

Sari,

Rachmawanto

et al. 2024

SJI

View full text Add to dashboard Cite

show abstract

“…The convolutional layers employ rectified linear unit (ReLU) activations, promoting faster convergence and better training efficiency. The final layers consist of fully connected layers with softmax activation, enabling classification into multiple classes [17] , as shown in Figure 1. VGG19 has been a pivotal model in the field of deep learning, showcasing the significance of depth in neural network architectures.…”

Section: A Vgg19 Modelmentioning

confidence: 99%

“…This approach entails using the network up to a predefined layer as an arbitrary feature extractor, with the outputs of these layers serving as features for further processing [28]. The VGG19 model, known for its 16 convolutional layers and three output layers, leverages its convolutional layers for feature extraction, preserving the original features of input images in the form of feature maps [17].…”

Section: ) Feature Extraction Approachmentioning

confidence: 99%

An Approach for Aircraft Detection using VGG19 and OCSVM

A. Hameed,

A. Khalaf

2024

IJCDS

View full text Add to dashboard Cite

Aircraft detection is an essential and noteworthy area of object detection that has received significant interest from scholars, especially with the progress of deep learning techniques. Aircraft detection is now extensively employed in various civil and military domains. Automated aircraft detection systems play a crucial role in preventing crashes, controlling airspace, and improving aviation traffic and safety on a civil scale. In the context of military operations, detection systems play a crucial role in quickly locating aircraft for surveillance purposes, enabling decisive military strategies in real time. This article proposes a system that accurately detects airplanes independent of their type, model, size, and color variations. However, the diversity of aircraft images, including variations in size, illumination, resolution, and other visual factors, poses challenges to detection performance. As a result, an aircraft detection system must be designed to distinguish airplanes clearly without affecting the aircraft's position, rotation, or visibility. The methodology involves three significant steps: feature extraction, detection, and evaluation. Firstly, deep features will be extracted using a pre-trained VGG19 model and transfer learning principle. Subsequently, the extracted feature vectors are employed in One Class Support Vector Machine (OCSVM) for detection purposes. Finally, the results are assessed using evaluation criteria to ensure the effectiveness and accuracy of the proposed system. The experimental evaluations were conducted across three distinct datasets: Caltech-101, Military dataset, and MTARSI dataset. Furthermore, the study compares its experimental results with those of comparable publications released in the past three years. The findings illustrate the efficacy of the proposed approach, achieving F1-scores of 96% on the Caltech-101 dataset and 99% on both Military and MTARSI datasets.

show abstract

“…Dunderdale et al [9] propose a feature-oriented, deep-learning strategy for the detection of defective PV cell surfaces. For the purpose of benchmarking, the dataset is trained on the VGG-16 [10] and MobileNet [11] architectures. Furthermore, the authors provide a comparison between the de facto ADAM optimizer against the stochastic gradient descent (SGD).…”

Section: Literature Reviewmentioning

confidence: 99%

PV-CrackNet Architecture for Filter Induced Augmentation and Micro-Cracks Detection within a Photovoltaic Manufacturing Facility

2022

View full text Add to dashboard Cite

Photovoltaic cell manufacturing is a rigorous process involving many stages where the cell surface is exposed to external pressure and temperature differentials. This provides fertile ground for micro-cracks to develop on the cell surface. At present, domain experts carry out a manual inspection of the cell surface to judge if any micro-cracks are present. This research looks to overcome the issue of cell data scarcity through the proposed filter-induced augmentations, thus providing developers with an effective, cost-free mechanism for generating representative data samples. Due to the abstract nature of the cell surfaces, the proposed augmentation strategy is effective in generating representative samples for better generalization. Furthermore, a custom architecture is developed that is computationally lightweight compared to state-of-the-art architectures, containing only 7.01 million learnable parameters while achieving an F1-score of 97%.

show abstract

Remote Sensing Data Classification Using A Hybrid Pre-Trained VGG16 CNN- SVM Classifier

Cited by 19 publications

References 6 publications

A Comparative Study of Javanese Script Classification with GoogleNet, DenseNet, ResNet, VGG16 and VGG19

A Comparative Study of Javanese Script Classification with GoogleNet, DenseNet, ResNet, VGG16 and VGG19

An Approach for Aircraft Detection using VGG19 and OCSVM

PV-CrackNet Architecture for Filter Induced Augmentation and Micro-Cracks Detection within a Photovoltaic Manufacturing Facility

Contact Info

Product

Resources

About