Optimizing Grape Leaf Disease Identification Through Transfer Learning and Hyperparameter Tuning

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(2)

Published: Jan. 1, 2024

Grapes are a globally cultivated fruit with significant economic and nutritional value, but they susceptible to diseases that can harm crop quality yield. Identifying grape leaf accurately promptly is vital for effective disease management sustainable viticulture. To address this challenge, we employ transfer learning approach, utilizing well-established pre-trained models such as ResNet50V2, ResNet152V2, MobileNetV2, Xception, In-ceptionV3, renowned their exceptional performance across various tasks. Our primary objective identify the most suitable network architecture classification of diseases. This achieved through rigorous evaluation process considers key metrics accuracy, F1 score, precision, recall, loss. By systematically assessing these models, aim select one demonstrates best on our dataset. Following model selection, proceed crucial phase fine-tuning model’s hyperparameters. essential enhance predictive capabilities overall effectiveness in identification. accomplish this, conduct an extensive hyperparameter search using Hyperband strategy. Hyperparameters play pivotal role shaping behavior deep by exploring wide range combinations, goal optimal configuration maximizes given Additionally, study’s results were compared those numerous relevant studies.

Language: Английский

Comparing hybrid models for recognising objects in thermal images at nighttime

Indonesian Journal of Electrical Engineering and Computer Science, Journal Year: 2024, Volume and Issue: 34(3), P. 1823 - 1823

Published: April 5, 2024

This research aims to revolutionize urban object recognition by developing cloud-based Python programs using intelligent algorithms. Unlike current models that focus on colour enhancement in nighttime thermal images, this work addresses the critical challenge of accurate detection landscapes. The proposed method incorporates a binary generative adversarial network (GAN) generator can switch bidirectionally between daytime (DC) and infrared (NTIR) images. memory-based visual image memory (MVAM), system extracts important descriptive information from landscape reducing problems related small sample sizes. discussion presents comprehensive improvement evaluation deep learning classification pipeline Google Colab, demonstrating advanced processing. Using TensorFlow, Keres scikit libraries combined with algorithms such as DenseNet121 MobileNetV2 clear approach. We created Bidirectional GAN + MVAM for work. Our performed well, an accuracy 81.43%, precision 51.16, recall 50.11, F-score 46.37. systematic presentation code careful strategy ensure optimal performance, stability, efficiency processing tasks.

Language: Английский

Securing Networks: An In-Depth Analysis of Intrusion Detection using Machine Learning and Model Explanations

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(5)

Published: Jan. 1, 2024

As cyber threats continue to evolve in complexity, the need for robust intrusion detection systems (IDS) becomes increasingly critical. Machine learning (ML) models have demon-strated their effectiveness detecting anomalies and potential intrusions. In this article, we delve into world of by exploring application four distinct ML models: XGBoost, Decision Trees, Random Forests, Bagging. And leveraging interpretability tools LIME (Local Interpretable Model-agnostic Explanations) SHAP (SHapley Additive ex-Planations) explain classification results. Our exploration begins with an in-depth analysis each machine model, shedding light on strengths, weaknesses, suitability detection. However, often operate as "black boxes" making it crucial inner workings. This article introduces indispensable model interpretability. Throughout demonstrate practical interpret output our models. By doing so, gain valuable insights decision-making process these models, enhancing ability identify respond effectively.

Language: Английский