IS&T | Library

10.2352/J.ImagingSci.Technol.2025.69.4.040101

Abstract

A collection of articles on remote research in cognition and perception using the Internet for the Journal of Perceptual Imaging is presented. Four original articles cover the topics of exact versus conceptual replication of cognitive effects (e.g., mental accounting), effects of facial cues on the perception of avatars, cultural influences on perceptual image and video quality assessment, and how Internet habits influence social cognition and social cognitive research. The essentials of these articles are summarized here, and their contributions are embedded within a wider view and historical perspective on remote research in cognition and perception using the Internet.

Digital Library: JPI

Published Online: August 2025

Editorial

4 2

From the Editor

Chunghui Kuo

Page 1, July 2025, © Society for Imaging Science and Technology 2025

DOI

Volume 69

Issue 4

Digital Library: JIST

Published Online: July 2025

Applying the Taguchi Method to the Optimized Design of YOLO Hyperparameters: Blood Cell Count and Detection

6 2

Taguchi method
hyperparameter
robust optimization
deep learning
object detection
YOLO

Fu-I Chou, Tian-Hsiang Huang, Po-Yuan Yang, Chun-Chieh Huang, Meng-Hsuan Chiang, Wen-Hsien Ho, Jyh-Horng Chou

Pages 1 - 8, July 2025, © Society for Imaging Science and Technology 2025

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040401

Volume 69

Issue 4

Abstract

In recent years, deep learning has achieved excellent results in several applications across various fields. However, as the scale of deep learning models increases, the training time of the models also increases dramatically. Furthermore, hyperparameters have a significant influence on model training results and selecting the model’s hyperparameters efficiently is essential. In this study, the orthogonal array of the Taguchi method is used to find the best experimental combination of hyperparameters. This research uses three hyperparameters of the you only look once-version 3 (YOLOv3) detector and five hyperparameters of data augmentation as the control factor of the Taguchi method in addition to the traditional signal-to-noise ratio (S/N ratio) analysis method with larger-the-better (LB) characteristics.

Experimental results show that the mean average precision of the blood cell count and detection dataset is 84.67%, which is better than the available results in literature. The method proposed herein can provide a fast and effective search strategy for optimizing hyperparameters in deep learning.

Digital Library: JIST

Published Online: July 2025

Hybrid Deep Learning Fusion Combined Technology Acceptance Models for Counterfeit Label Classification and Satisfaction Assessment Applications

7 2

anti-counterfeit label classification
transfer learning
neighborhood components analysis
Naive Bayes
support vector machine
PLS-SEM

Hua-Ching Chen, Hsuan-Ming Feng

Pages 1 - 19, July 2025, © Society for Imaging Science and Technology 2025

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040402

Volume 69

Issue 4

Abstract

This article resolutely uses the concept of feature fusion to establish a deep learning model that can quickly recognize objects and complete an anti-counterfeit label recognition system. The receiver combines the training of the technology acceptance model (TAM) to evaluate the satisfaction of users in completing the anti-counterfeit label classification training. In this study, the fusion-based recognition program was employed to extract the feature sets of different categories of anti-counterfeit labels based on the operation of multilayer convolutional neural networks (CNNs) with different depth models. Using neighborhood components analysis, ten important sets of features from different CNN models were selected and reorganized parallelly into a new small-scale feature fusion dataset. By using naive Bayes and support vector machine methods, efficient classification of wine label image feature datasets after fusion was achieved. The feature fusion anti-counterfeiting label recognition system proposed in this article had a maximum recognition accuracy of 99.29% and a data reduction compression ratio of about 1/50. In addition to reducing training time, it maintained a high level of accuracy. This study established a TAM with the advantage of a feature fusion anti-counterfeit label recognition system. The model was tested on 100 consumers, and a satisfaction evaluation and validation analysis with partial least squares structural equation modeling were completed thereafter. The efficiency of the fusion-based deep learning model met the level of consumer satisfaction. This will be beneficial for educating consumers to use and enhance their willingness to promote and repurchase wine products in the future.

Digital Library: JIST

Published Online: July 2025

Single-Shot Detector with Resnet50 Backbone for Detecting Two Types of Colonic Polyps in Endoscopic Images

6 2

colonic polyps
adenoma
hyperplastic polyp
single-shot detector
Resnet50 network
Taguchi method

Yao-Mei Chen, Tak-Kee Choy, Wei-Tai Huang, Meng-Hsuan Chiang, Jinn-Tsong Tsai, Wen-Hsien Ho

Pages 1 - 12, July 2025, © Society for Imaging Science and Technology 2025

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040403

Volume 69

Issue 4

Abstract

Physicians attempt to detect different colonic polyps at the same time during endoscopy inspection. A deep-learning-based object detection method is proposed to aim at the problem of simultaneous detection of different colonic polyps. This study used a single-shot detector (SSD) with a Resnet50 backbone, called the SSD-Resnet50 model, to detect two types of colonic polyps, which are adenomas and hyperplastic polyps, in endoscopic images. The Taguchi method was used to optimize algorithm hyperparameter combinations for the SSD-Resnet50 model to promote the detection accuracy of colonic polyps. The SSD-Resnet50 model along with its optimized algorithm hyperparameters was employed for simultaneous detection of two types of colonic polyps. The experimental findings revealed that the SSD-Resnet50 model achieved an average mAP of 0.8933 on a test set comprising 300 × 300 × 3 images of colonic polyps. Notably, the detection accuracy attained with the SSD-Resnet50 model and its optimized algorithm hyperparameters, derived from the Taguchi method, surpassed that of the SSD-Resnet50 model and its algorithm hyperparameter combination obtained from the Matlab example. Additionally, the SSD-Resnet50 model achieved higher detection accuracy compared to the SSD-MobileNetV2, SSD-InceptionV3, SSD-Shufflenet, SSD-Squeezenet, and SSD-VGG16 models. The proposed SSD-Resnet50 model with its optimized algorithm hyperparameters had higher accuracy in detecting the adenomas and hyperplastic polyps in endoscopic images at the same time.

Digital Library: JIST

Published Online: July 2025

Optimized Multi-Class Urban Appearance Violation Detection Model in UAV Imagery

7 2

urban appearance violation
UAV imagery
custom dataset
object detection
attention mechanism

Songlin Wei, Hua-Ching Chen, Hsuan-Ming Feng, Weiquan Li

Pages 1 - 9, July 2025, © Society for Imaging Science and Technology 2025

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040404

Volume 69

Issue 4

Abstract

The detection of urban appearance violation in unmanned aerial vehicle imagery faces several challenges. To address this problem, an optimized YOLOv8n-based urban appearance violation detection model is proposed. A custom dataset including four classes is created owing to the lack of a sufficient dataset. The Convolutional Block Attention Module attention mechanism is applied to improve the feature extraction ability of the model. A small target detection head is added to capture the characteristics of small targets and context information more effectively. The loss function Wise Intersection over Union is applied to improve the regression performance of the bounding box and the robustness of detection. Experimental results show that compared with the YOLOv8n model, the Precision, Recall, mAP_0.5, and mAP_0.5−0.95 of the optimized method increase by 3.8%, 2.1%, 3.3%, and 4.8%, respectively. Besides, an intelligent urban appearance violation detection system is developed, which generates and delivers warning messages via the WeChat official account platform.

Digital Library: JIST

Published Online: July 2025

Application of YOLOv7-tiny for Guava Covering Status Detection and Quantity Estimation

4 2

agricultural automation
guava
covering detection
YOLOv7-tiny

Dyi-Cheng Chen, Shang-Wei Lu, Li-Chan Lu

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040405

Volume 69

Issue 4

Work Presented at Electronic Imaging 2025

Abstract

This study applies YOLOv7-tiny object detection to inspect guava covering and count their quantity. Real-time monitoring enhances efficiency and reduces labor costs in agriculture. A custom dataset was created by collecting and labeling guava images. The YOLOv7-tiny model, trained with default parameters, achieved an initial mean Average Precision (mAP) of 66.7%. To improve accuracy, parameter adjustments, data augmentation (mosaic, mixup), and learning rate strategies (warm-up, decay) were employed, raising the mAP to 76.7%. The optimized model was transferred to mobile devices for convenient detection. This research provides an effective method for guava covering inspection and quantity counting, contributing to advancements in agricultural applications.

Digital Library: JIST

Published Online: July 2025

5 2

ExtremeMETA: High-speed Lightweight Image Segmentation Model by Remodeling Multi-channel Metamaterial Imagers

large convolution kernel
model compression
segmentation
meta-material

Quan Liu, Brandon T. Swartz, Ivan Kravchenko, Jason G. Valentine, Yuankai Huo

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040406

Volume 69

Issue 4

Enhancing Compound Emotion Recognition via Ensemble Representation Learning

Abstract

Deep neural networks (DNNs) have heavily relied on traditional computational units, such as CPUs and GPUs. However, this conventional approach brings significant computational burden, latency issues, and high power consumption, limiting their effectiveness. This has sparked the need for lightweight networks such as ExtremeC3Net. Meanwhile, there have been notable advancements in optical computational units, particularly with metamaterials, offering the exciting prospect of energy-efficient neural networks operating at the speed of light. Yet, the digital design of metamaterial neural networks (MNNs) faces precision, noise, and bandwidth challenges, limiting their application to intuitive tasks and low-resolution images. In this study, we proposed a large kernel lightweight segmentation model, ExtremeMETA. Based on ExtremeC3Net, our proposed model, ExtremeMETA maximized the ability of the first convolution layer by exploring a larger convolution kernel and multiple processing paths. With the large kernel convolution model, we extended the optic neural network application boundary to the segmentation task. To further lighten the computation burden of the digital processing part, a set of model compression methods was applied to improve model efficiency in the inference stage. The experimental results on three publicly available datasets demonstrated that the optimized efficient design improved segmentation performance from 92.45 to 95.97 on mIoU while reducing computational FLOPs from 461.07 MMacs to 166.03 MMacs. The large kernel lightweight model ExtremeMETA showcased the hybrid design’s ability on complex tasks.

Digital Library: JIST

Published Online: July 2025

Regular Article

4 2

deep neural networks
ensemble model
compound emotion recognition
deep learning

Sana Ullah, Guangyao Zhou, Jie Ou, Zhaokun Wang, Wenhong Tian

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040501

Volume 69

Issue 4

Federated Learning for Image-based Cancer Detection and Diagnosis: A Systematic Review

Abstract

Innovations in computer vision have steered research towards recognizing compound facial emotions, a complex mix of basic emotions. Despite significant advancements in deep convolutional neural networks improving accuracy, their inherent limitations, such as gradient vanishing/exploding problem, lack of global contextual information, and overfitting issues, may degrade performance or cause misclassification when processing complex emotion features. This study proposes an ensemble method in which three pre-trained models, DenseNet-121, VGG-16, and ResNet-18 are concatenated instead of utilizing individual models. It is a significant layer-sharing method, and we have added dropout layers, fully connected layers, activation functions, and pooling layers to each model after removing their heads before concatenating them. This enables the model to get a chance to learn more before combining the individual learned features. The proposed model uses an early stopping mechanism to prevent it from overfitting and improve performance. The proposed ensemble method surpassed the state-of-the-art (SOTA) with 74.4% and 71.8% accuracy on RAF-DB and CFEE datasets, respectively, offering a new benchmark for real-world compound emotion recognition research.

Digital Library: JIST

Published Online: July 2025

Regular Article

4 2

cancer
computer-aided detection and diagnosis systems
deep learning
federated learning
privacy-preserving

Ye Jin, Jinshan Tang

DOI

10.2352/J.ImagingSci.Technol.2025.69.4.040502

Volume 69

Issue 4