Guoxiang Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Guoxiang Huang has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Supapan Chaiprapat
Department of Industrial and Manufacturing Engineering, Faculty of Engineering, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand
4 shared publications - Kunlapat Thongkaew
Department of Industrial and Manufacturing Engineering, Faculty of Engineering, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand
3 shared publications - Nutt Nuntapong
Aquatic Science and Innovative Management Division, Faculty of Natural Resources, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand
2 shared publications Kriangkrai Waiyagan
Prince of Songkla University
1 shared publication
Latest Publications
- Open Access
Advancing Non-Destructive Ovarian Maturity Classification in Live Female Mud Crabs Through Deep Learning and Expert Visual Assessment
Reliable ovarian maturity assessment can support harvest decisions in cultured female mud crabs (Scylla paramamosain), but evaluating internal ovarian development non-destructively remains challenging. This study developed and evaluated a convolutional neural network (CNN)-based framework for binary ovarian maturity classification from smartphone-acquired RGB images without handcrafted feature extraction. Classification performance was systematically compared across two imaging modes: dorsal transillumination imaging (DTI) and flashlight-assisted imaging (FAI). The dataset comprised 2000 images (1000 per mode), with reference labels assigned by expert consensus. Five CNN architectures were evaluated under matched-condition, cross-condition, and combined-development settings, with five runs per architecture and development setting. Under matched conditions, VGG16 achieved the highest mean accuracy for DTI (0.9933), whereas DenseNet121 achieved the highest for FAI (0.9893). Cross-condition accuracies ranged from 0.7187 to 0.8947 for DTI→FAI and from 0.6907 to 0.8067 for FAI→DTI. Under combined DTI+FAI development, the highest mean accuracies were 0.9920 on DTI (VGG16) and 0.9947 on FAI (MobileNetV2). Combined development yielded higher mean accuracy than cross-condition evaluation on the same test subset for all five architectures. These findings characterize CNN performance across the evaluated imaging modes and may inform future non-destructive ovarian maturity assessment tools.
Open AccessAdvancing Non-Destructive Ovarian Maturity Classification in Live Female Mud Crabs Through Deep Learning and Expert Visual Assessment
Reliable ovarian maturity assessment can support harvest decisions in cultured female mud crabs (Scylla paramamosain), but evaluating internal ovarian development non-destructively remains challenging. This study developed and evaluated a convolutional neural network (CNN)-based framework for binary ovarian maturity classification from smartphone-acquired RGB images without handcrafted feature extraction. Classification performance was systematically compared across two imaging modes: dorsal transillumination imaging (DTI) and flashlight-assisted imaging (FAI). The dataset comprised 2000 images (1000 per mode), with reference labels assigned by expert consensus. Five CNN architectures were evaluated under matched-condition, cross-condition, and combined-development settings, with five runs per architecture and development setting. Under matched conditions, VGG16 achieved the highest mean accuracy for DTI (0.9933), whereas DenseNet121 achieved the highest for FAI (0.9893). Cross-condition accuracies ranged from 0.7187 to 0.8947 for DTI→FAI and from 0.6907 to 0.8067 for FAI→DTI. Under combined DTI+FAI development, the highest mean accuracies were 0.9920 on DTI (VGG16) and 0.9947 on FAI (MobileNetV2). Combined development yielded higher mean accuracy than cross-condition evaluation on the same test subset for all five architectures. These findings characterize CNN performance across the evaluated imaging modes and may inform future non-destructive ovarian maturity assessment tools.
- Open Access
Toward Sustainable Aquaculture: An Image-Based Framework for Ovarian Maturity Assessment in Live Female Mud Crabs
Ovarian maturity in live female mud crabs (Scylla paramamosain) strongly affects harvest decisions and market value. Current ovarian maturity assessment relies mainly on expert-dependent methods that are subjective and destructive. Therefore, this study aimed to develop an interpretable, non-destructive image-based framework to classify crab ovarian maturity as immature or mature. A total of 240 crab image sets acquired using ventral external-view, dorsal external-view, and dorsal transillumination imaging were retained for analysis. Six primary morphometric features were semi-manually extracted from these views. External-view images quantified carapace width (CW), abdomen width (AW), abdomen area (AA), and sternum area (SA). Dorsal transillumination images yielded carapace area (CA) and ovary area (OA), an internal cue visualized through the intact carapace. To mitigate body-size variation, three ratio-based features—abdomen–carapace width ratio (ACWR), abdomen–sternum area ratio (ASAR), and ovary–carapace area ratio (OCAR)—were calculated. Between-class comparisons and correlation analyses were performed to guide candidate feature-set construction. Because OA and OCAR were strongly correlated, two reduced feature sets (Reduced 1 and Reduced 2) were designed to compare absolute ovary area with normalized ovary occupancy. Five feature sets—Raw, Ratio, Combined, Reduced 1, and Reduced 2—were evaluated using logistic regression (LR), support vector machine (SVM), and random forest (RF) classifiers. The Combined feature set, integrating all primary and ratio-based features, achieved the strongest mean cross-validated performance when paired with LR. On the held-out test set (n = 40), the final Combined-LR model achieved 0.950 accuracy and 0.997 ROC–AUC. On an independent practical implementation set (n = 40), the model correctly classified 39 specimens, achieving 0.975 accuracy. These findings may support non-destructive ovarian maturity screening and commercial grading in mud crab aquaculture.
Open AccessToward Sustainable Aquaculture: An Image-Based Framework for Ovarian Maturity Assessment in Live Female Mud Crabs
Ovarian maturity in live female mud crabs (Scylla paramamosain) strongly affects harvest decisions and market value. Current ovarian maturity assessment relies mainly on expert-dependent methods that are subjective and destructive. Therefore, this study aimed to develop an interpretable, non-destructive image-based framework to classify crab ovarian maturity as immature or mature. A total of 240 crab image sets acquired using ventral external-view, dorsal external-view, and dorsal transillumination imaging were retained for analysis. Six primary morphometric features were semi-manually extracted from these views. External-view images quantified carapace width (CW), abdomen width (AW), abdomen area (AA), and sternum area (SA). Dorsal transillumination images yielded carapace area (CA) and ovary area (OA), an internal cue visualized through the intact carapace. To mitigate body-size variation, three ratio-based features—abdomen–carapace width ratio (ACWR), abdomen–sternum area ratio (ASAR), and ovary–carapace area ratio (OCAR)—were calculated. Between-class comparisons and correlation analyses were performed to guide candidate feature-set construction. Because OA and OCAR were strongly correlated, two reduced feature sets (Reduced 1 and Reduced 2) were designed to compare absolute ovary area with normalized ovary occupancy. Five feature sets—Raw, Ratio, Combined, Reduced 1, and Reduced 2—were evaluated using logistic regression (LR), support vector machine (SVM), and random forest (RF) classifiers. The Combined feature set, integrating all primary and ratio-based features, achieved the strongest mean cross-validated performance when paired with LR. On the held-out test set (n = 40), the final Combined-LR model achieved 0.950 accuracy and 0.997 ROC–AUC. On an independent practical implementation set (n = 40), the model correctly classified 39 specimens, achieving 0.975 accuracy. These findings may support non-destructive ovarian maturity screening and commercial grading in mud crab aquaculture.
- Open Access
- 30 January 2026
- Published by MDPI AG in Aquaculture Journal
- Vol. 6 (1) , 3
A Review of Non-Destructive Technologies for Quality Assessment in Aquaculture
Aquatic animal products are vital to global food security and nutrition, necessitating accurate, scalable, and non-destructive methods for quality assessment in aquaculture. Conventional techniques such as dissection and biochemical analysis are invasive, labor-intensive, and unsuitable for real-time or high-throughput decision-making. This review synthesizes six major categories of non-destructive technologies—electrical, spectroscopic, natural sensory, acoustic, radiographic, and infrared and microwave—classified by their underlying sensing mechanisms and therefore differing measurement capabilities and deployment feasibilities. To support objective technology selection, an Analytic Hierarchy Process (AHP) framework was developed using general performance criteria (cost, accuracy, speed, usability) and one decision-critical application-specific criterion (non-invasiveness), and was demonstrated for ovarian maturation staging in mud crabs by ranking 19 candidate techniques. Accuracy had the highest weight (0.416), but non-invasiveness (0.224) and usability (0.197) substantially influenced the final ranking, illustrating how operational and welfare constraints could shift preferred solutions despite differences in analytical accuracy. Based on the global priority weights (GA), computer vision (CV) was identified as the most suitable option (GA = 0.076), balancing affordability, throughput, ease of deployment, and animal welfare compatibility, whereas high-end modalities such as nuclear magnetic resonance (NMR; GA = 0.073) and computed tomography (CT; GA = 0.070) were constrained by cost and operational complexity. Overall, this review–AHP–case study pipeline provides a transparent and reproducible decision-support basis for selecting non-destructive technologies across aquaculture species and quality targets.
- 30 January 2026
- Published by MDPI AG in Aquaculture Journal
- Vol. 6 (1) , 3
Open AccessA Review of Non-Destructive Technologies for Quality Assessment in Aquaculture
Aquatic animal products are vital to global food security and nutrition, necessitating accurate, scalable, and non-destructive methods for quality assessment in aquaculture. Conventional techniques such as dissection and biochemical analysis are invasive, labor-intensive, and unsuitable for real-time or high-throughput decision-making. This review synthesizes six major categories of non-destructive technologies—electrical, spectroscopic, natural sensory, acoustic, radiographic, and infrared and microwave—classified by their underlying sensing mechanisms and therefore differing measurement capabilities and deployment feasibilities. To support objective technology selection, an Analytic Hierarchy Process (AHP) framework was developed using general performance criteria (cost, accuracy, speed, usability) and one decision-critical application-specific criterion (non-invasiveness), and was demonstrated for ovarian maturation staging in mud crabs by ranking 19 candidate techniques. Accuracy had the highest weight (0.416), but non-invasiveness (0.224) and usability (0.197) substantially influenced the final ranking, illustrating how operational and welfare constraints could shift preferred solutions despite differences in analytical accuracy. Based on the global priority weights (GA), computer vision (CV) was identified as the most suitable option (GA = 0.076), balancing affordability, throughput, ease of deployment, and animal welfare compatibility, whereas high-end modalities such as nuclear magnetic resonance (NMR; GA = 0.073) and computed tomography (CT; GA = 0.070) were constrained by cost and operational complexity. Overall, this review–AHP–case study pipeline provides a transparent and reproducible decision-support basis for selecting non-destructive technologies across aquaculture species and quality targets.
- 9 December 2015
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 86 (1-4) , 323-335
Automated process planning and cost estimation under material quality uncertainty
- 9 December 2015
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 86 (1-4) , 323-335
Automated process planning and cost estimation under material quality uncertainty