Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

FINE-TUNING MobileNetV3 WITH DIFFERENT WEIGHT OPTIMIZATION ALGORITHMS FOR CLASSIFICATION OF DENOISED BLOOD CELL IMAGES USING CONVOLUTIONAL NEURAL NETWORK

Domain:

healthcare

Record type:

paper
Creator:
M. N.
Publisher:
Beg
Host:
Breast cancer remains a formidable global health concern, underscoring the urgency for advanced diagnostic methodologies. This research presents a multifaceted framework aimed at significantly enhancing breast cancer diagnosis through innovative approaches in image processing and machine learning. The proposed framework encompasses several key contributions. Firstly, a robust denoising strategy is implemented using Convolutional Neural Network encoder-decoder architecture, augmented with data augmentation techniques. This addresses the challenge of vanishing gradients through enhanced Rectified Linear Units based Convolutional Neural Network, enhancing the model's generalization capability. Subsequent to denoising, feature extraction is performed utilizing a fine-tuned MobileNetV3 model. The model's performance is optimized through Modified Rectified Linear Units and NRMSProp approaches, effectively eliminating undesired features and improving overall efficiency. Crucially, a novel feature selection process is introduced, leveraging the Artificial Hummingbird Algorithm based on Manta Ray Foraging Optimization Algorithm. This algorithm selectively identifies essential features from breast cancer images, significantly elevating classification accuracy. To validate the proposed framework, a comprehensive evaluation is conducted, comparing its performance with a hybrid of five different metaheuristic algorithms, including Marine Predators Algorithm, Tunicate Swarm Algorithm, Manta Ray Foraging Optimization algorithm, Arithmetic Optimization Algorithm, and Jelly Fish optimization algorithm. Artificial Hummingbird Algorithm based on Manta Ray Foraging Optimization Algorithm emerges as the most effective among these algorithms, showcasing superior performance. The evaluation utilized the Breast Cancer Histopathological Image Classification dataset, resulting in an impressive classification accuracy of 99.51% for the proposed model.

Visit

doi.org

Tasks

computer visionimage classification

Languages

Manta

Similar

Transfer Learning Using Convolutional Neural Network Architectures for Brain Tumor Classification from MRI ImagesClassification of COVID-19 from CT chest images using Convolutional Wavelet Neural NetworkLight-Weight Deep Convolutional Neural Network Model for Classification of Potato Leaf DiseasesMulti-Scale Classification of Sentinel-2 Images for Land Cover Mapping Using Two-Branch Convolutional Neural NetworkLung Cancer Classification Based on CT Images Using Hybrid Convolutional Neural Network-Random Forest ModelArtificial and Convolutional Neural Network Architectures for Childhood Stunting Classification: Design, Evaluation, and Optimization

Transfer Learning Using Convolutional Neural Network Architectures for Brain Tumor Classification from MRI Images

Part 3: Image processing International audience Brain tumor classification is very im

Classification of COVID-19 from CT chest images using Convolutional Wavelet Neural Network

Analyzing X-rays and computed tomography-scan (CT scan) images using a convolutional neural

Light-Weight Deep Convolutional Neural Network Model for Classification of Potato Leaf Diseases

Potato leaf diseases pose a significant threat to global food security by reducing crop yields and e

Multi-Scale Classification of Sentinel-2 Images for Land Cover Mapping Using Two-Branch Convolutional Neural Network

International audience Effectively characterize the current land cover status is cruc

Lung Cancer Classification Based on CT Images Using Hybrid Convolutional Neural Network-Random Forest Model

International audience

Lung cancer is a type of cancer that starts when abnor

Artificial and Convolutional Neural Network Architectures for Childhood Stunting Classification: Design, Evaluation, and Optimization

Childhood stunting is a global health challenge, affecting 148 million children under 5 in 2022. It