Classification of Malaria Parasite Plasmodium Falciparum Based on Blood Smear Images Using Support Vector Machine Approach DOI
Nur Chamidah, Toha Saifudin, Riries Rulaningtyas

et al.

Data & Metadata, Journal Year: 2024, Volume and Issue: 4, P. 568 - 568

Published: Dec. 12, 2024

Malaria remained a significant global health issue, particularly in tropical and subtropical regions. The disease resulted substantial number of clinical cases deaths each year, with high-risk groups including infants, toddlers, pregnant women. Accurate prompt diagnosis was key factor managing the disease. To address this research aimed to develop an automated system for classification Plasmodium falciparum malaria parasites based on blood smear images. methods employed included image feature selection using Principal Component Analysis (PCA) Support Vector Machine (SVM) approach classification. findings indicated that process, category normal exhibited distinctive characteristics PC1 PC2 values tended be negative dispersed, whereas parasitic displayed greater variability both components. Furthermore, evaluation system's accuracy SVM three different kernel types showed promising results. average through K-fold cross-validation polyinomial, linear, radial basis function kernels 96.7%, 98.9%, 94.4%, respectively. These results highlighted potential utilization

Language: Английский

CNNM-FDI: Novel Convolutional Neural Network Model for Fire Detection in Images DOI
Arvind Kumar Vishwakarma, Maroti Deshmukh

IETE Journal of Research, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 14

Published: Jan. 29, 2025

Language: Английский

Citations

1
Computer Vision-Based Concrete Crack Identification Using MobileNetV2 Neural Network and Adaptive Thresholding DOI Creative Commons
Li Hui, Ahmed Ibrahim,

Riyadh Hindi

et al.

Infrastructures, Journal Year: 2025, Volume and Issue: 10(2), P. 42 - 42

Published: Feb. 18, 2025

Concrete is widely used in different types of buildings and bridges; however, one the major issues for concrete structures crack formation propagation during its service life. These cracks can potentially introduce harmful agents into concrete, resulting a reduction overall lifespan structures. Traditional methods detection primarily hinge on manual visual inspection, which relies experience expertise inspectors using tools such as magnifying glasses microscopes. To address this issue, computer vision most innovative solutions cracking evaluation, application has been an area research interest past few years. This study focuses utilization lightweight MobileNetV2 neural network detection. A dataset including 40,000 images was adopted preprocessed various thresholding techniques, adaptive selected developing evaluation algorithm. While both convolutional (CNN) indicated comparable accuracy levels detection, model’s significantly smaller size makes it more efficient selection mobile devices. In addition, advanced algorithm developed to detect evaluate widths high-resolution images. The effectiveness reliability method were subsequently assessed through experimental validation.

Language: Английский

Citations

1
Grey prediction evolution algorithm with a dominator guidance strategy for solving multi-level image thresholding DOI
Peixin Yang, Zhongbo Hu, Yang Zhou

et al.

Applied Soft Computing, Journal Year: 2025, Volume and Issue: unknown, P. 112947 - 112947

Published: March 1, 2025

Language: Английский

Citations

0
A control-driven transition strategy for enhanced multi-level threshold image segmentation optimization DOI
Laith Abualigah,

Mohammad H. Almomani,

Saleh Ali Alomari

et al.

Egyptian Informatics Journal, Journal Year: 2025, Volume and Issue: 30, P. 100646 - 100646

Published: April 11, 2025

Language: Английский

Citations

0
Dynamic comprehensive learning-based dung beetle optimizer using triangular mutation for polyps image segmentation DOI
Mohamed Abd Elaziz, Diego Oliva,

Alaa A. El‐Bary

et al.

Computational Biology and Chemistry, Journal Year: 2025, Volume and Issue: 118, P. 108474 - 108474

Published: April 24, 2025

Language: Английский

Citations

0
Dynamic display algorithm of sonar data based on grayscale distribution model and computational intelligence DOI Creative Commons

Hongquan Lei,

Diquan Li,

Haidong Jiang

et al.

Frontiers in Earth Science, Journal Year: 2025, Volume and Issue: 13

Published: April 28, 2025

Existing image processing and target recognition algorithms have limitations in complex underwater environments dynamic changes, making it difficult to ensure real-time precision. Multiple noise sources interfere with sonar signals, which affects both data precision clarity. This article studies the display algorithm of based on grayscale distribution model computational intelligence. It proposes construct a for images, analyze histogram, determine threshold selection maximum entropy segmentation method, finally complete segmentation. The segmented images can be used train convolutional neural network object constructed this article. To verify effectiveness proposed test set was evaluate trained model. 87.95%, recall 87.97%, F1 value 0.8794, is significantly higher than traditional (Such as Otsu SVM below 80%). speed reached 37 m, certain improvement compared

Language: Английский

Citations

0
Disease Monitoring and Characterization of Feeder Road Network Based on Improved YOLOv11 DOI Open Access
Ying Fan, Kun Zhi, Haichao An

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(9), P. 1818 - 1818

Published: April 29, 2025

In response to the challenges of low accuracy and high misdetection omission rate disease detection on feeder roads, an improved Rural-YOLO (SAConv-C2f+C2PSA_CAA+MCSAttention+WIOU) algorithm is proposed in this paper, which enhanced target framework based YOLOv11 architecture, for identification common diseases complex road environment. The methodology introduces four key innovations: (1) Switchable Atrous Convolution (SAConv) introduced into backbone network enhance multiscale feature extraction under occlusion conditions; (2) Multi-Channel Spatial Attention (MCSAttention) constructed fusion process, weight distribution adjusted through adaptive redistribution. By adjusting distribution, model’s sensitivity subtle features improved. To its ability discriminate between different types, Cross Stage Partial with Parallel Channel Adaptive Aggregation (C2PSA_CAA) at end network. (3) mitigate category imbalance issues, Weighted Intersection over Union loss (WIoU_loss) introduced, helps optimize bounding box regression process improve relevant diseases. Based experimental validation, demonstrated superior performance minimal computational overhead. Only 0.7 M additional parameters required, 8.4% improvement recall a 7.8% increase mAP50 were achieved compared initial models. optimized architecture also reduced model size by 21%. test results showed that 3.28 complexity 5.0 GFLOPs, meeting requirements lightweight deployment scenarios. Cross-validation multi-scenario public datasets was carried out, robustness across diverse conditions. quantitative experiments, center skeleton method maximum internal tangent circle used calculate crack width, pixel occupancy ratio assess area damage degree potholes other measurements converted actual physical dimensions using calibrated scale 0.081:1.

Language: Английский

Citations

0
Joint Image Processing with Learning-Driven Data Representation and Model Behavior for Non-Intrusive Anemia Diagnosis in Pediatric Patients DOI Creative Commons
Tarek Berghout

Journal of Imaging, Journal Year: 2024, Volume and Issue: 10(10), P. 245 - 245

Published: Oct. 2, 2024

Anemia diagnosis is crucial for pediatric patients due to its impact on growth and development. Traditional methods, like blood tests, are effective but pose challenges, such as discomfort, infection risk, frequent monitoring difficulties, underscoring the need non-intrusive diagnostic methods. In light of this, this study proposes a novel method that combines image processing with learning-driven data representation model behavior anemia in patients. The contributions threefold. First, it uses an image-processing pipeline extract 181 features from 13 categories, feature-selection process identifying most learning. Second, deep multilayered network based long short-term memory (LSTM) utilized train classifying images into anemic non-anemic cases, where hyperparameters optimized using Bayesian approaches. Third, trained LSTM integrated layer learning developed recurrent expansion rules, forming part new called (RexNet). RexNet designed learn representations akin traditional deep-learning methods while also understanding interaction between dependent independent variables. proposed approach applied three public datasets, namely conjunctival eye images, palmar fingernail children aged up 6 years. achieves overall evaluation 99.83 ± 0.02% across all classification metrics, demonstrating significant improvements results generalization compared networks existing This highlights RexNet's potential promising alternative blood-based diagnosis.

Language: Английский

Citations

3
Fractional order calculus enhanced dung beetle optimizer for function global optimization and multilevel threshold medical image segmentation DOI
Huangzhi Xia, Yifen Ke,

Riwei Liao

et al.

The Journal of Supercomputing, Journal Year: 2024, Volume and Issue: 81(1)

Published: Oct. 28, 2024

Language: Английский

Citations

3
Multi-Level Image Segmentation Combining Chaotic Initialized Chimp Optimization Algorithm and Cauchy Mutation DOI Open Access
Shujing Li,

Zhangfei Li,

Wenhui Cheng

et al.

Computers, materials & continua/Computers, materials & continua (Print), Journal Year: 2024, Volume and Issue: 80(2), P. 2049 - 2063

Published: Jan. 1, 2024

To enhance the diversity and distribution uniformity of initial population, as well to avoid local extrema in Chimp Optimization Algorithm (CHOA), this paper improves CHOA based on chaos initialization Cauchy mutation. First, Sin is introduced improve random population scheme CHOA, which not only guarantees but also enhances population. Next, mutation added optimize global search ability process position (threshold) updating falling into optima. Finally, an improved was formed through combination (CICMCHOA), then taking fuzzy Kapur objective function, applied CICMCHOA natural medical image segmentation, compared it with four algorithms, including Satin Bowerbird optimizer (ISBO), Cuckoo Search (ICS), etc. The experimental results deriving from visual specific indicators demonstrate that delivers superior segmentation effects segmentation.

Language: Английский

Citations

1