The Journal of Spatial Innovation Development

An Analysis of User Needs Factors for Postal Services at Division of General Affairs, University of Phayao

Krisana Thungluang — 2026-08-07

Postal Services at Division of General Affairs, University of Phayao responsible for the delivery and receipt of documents and parcels across various departments. The service is supported by personnel and the "UP Postal" system. Previously, user needs had never been systematically surveyed or analyzed resulting in a lack of targeted improvements to meet their expectations. This research aimed to investigate the factors influencing user needs and to compare the importance placed on various service aspects by different user groups. In this study, the data was collected from 73 service users via questionnaires and evaluated using descriptive and non-parametric statistics. The findings revealed that users considered all factors to be of high importance, especially the use of the UP Postal system and service outcomes. The most critical factor in the system aspect was its user-friendliness and efficiency, while in the service outcome aspect, it was the overall positive impression of the service. Age-based comparisons revealed that users aged 35 and above placed significantly greater importance on the UP Postal system and service facilities than those aged 18–24 (p<0.05). Support staff considered the search function and facilities significantly more important than students (p<0.05). Therefore, users who both sent and received mail rated all service aspects significantly higher than those who only received mail (p<0.05). These findings can be applied to improve and develop the postal services in alignment with the specific needs of different user groups.

Lung Cancer Prediction from CT Scan Images Using Convolutional Neural Network

Jirayu Thongkam — 2026-08-07

Lung cancer is a serious disease and a leading cause of death both in Thailand and worldwide. Early detection is crucial, as it significantly increases the chances of successful treatment and reduces mortality rates. This research aims to study and develop a deep neural network model for lung cancer prediction using computed tomography scan images. The dataset consists of 1,000 images classified into four categories: adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and normal lung.

The study compared the performance of four deep learning models ResNet50, VGG16, InceptionV3, and DenseNet121. The results from each model were then combined using the Stacking Ensemble method with 5 machine learning techniques as meta models: Random Forest, Logistic Regression, Support Vector Machine, K-Nearest Neighbors, and Naive Bayes. Experimental results showed that the ensemble model using Random Forest achieved the highest accuracy of 98.73 percent, outperforming all individual models.

Design and Development of a Navigation System for the Visually Impaired Using the Internet of Things

Wirunchana Aounsri — 2026-08-07

Visually impaired individuals often face difficulties navigating public spaces with various obstacles, which can pose safety risks. This research aims to design and develop a navigation system for visually impaired persons to enhance safety and promote independence when walking on sidewalks. A compact and portable prototype device was developed in the form of a belt. The system operates by applying the YOLOv11n object detection algorithm to identify objects in the walking path, determine their positions, and provide appropriate avoidance guidance. It also integrates Internet of Things (IoT) technology, comprising a Raspberry Pi 5, a camera module, an ultrasonic sensor, and an MPU6050 sensor to detecting ground slope, along with a voice alert system. The implementation process consists of (1) Designing and developing the device, (2) Training and optimizing the YOLO model for real-world environmental conditions, and (3) Testing the performance of each device function.

Experimental results evaluating the performance of each function show that the system achieves an average object detection accuracy of 94.0%, an average avoidance guidance accuracy of 95.5%, and a slope detection accuracy of 89.0%. This research presents a potential approach for practical applications to assist visually impaired individuals.

Applying of Software for Automatic Seed Counting from Photographic Images

Witthaya Boonsuk — 2026-08-07

The objective of this research is to develop software for automatically counting seeds from photographic images. The system's performance was evaluated using image processing with the newly developed software on five sample image groups, each containing 100 seeds, totaling 500 seeds. The images used were in a standard resolution of 640x480 pixels. The accuracy of the system for each group was as follows: Group 1: 85% accuracy (average) Group 2: 95% accuracy (average) Group 3: 95% accuracy (average) Group 4: 95% accuracy (average) Group 5: 95% accuracy (average) Overall average accuracy: 93% To use the system, a camera must be connected to a computer to capture images directly, or pre-captured images from a standard camera can also be processed by the software.

Overall, the developed system is considered highly efficient. The accuracy of the software in counting seeds from images is relatively high, making it suitable for practical applications in automatic seed counting from photographic images.

Development of Online System for Preparation of Instrument and Equipment for Laboratory Teaching at Division of Microbiology and Parasitology, School of Medical Sciences, University of Phayao

Wasana Muangwong — 2026-08-07

The preparation of scientific equipment and materials for practical instruction in microbiology and parasitology is a critical factor influencing the effectiveness of the teaching and learning process. Traditional preparation systems often encounter issues related to coordination, data accuracy, and operational timelines, which may negatively impact the overall quality of instruction. This research aims to develop a data recording system for the preparation of scientific equipment and materials used in laboratory teaching, and to assess user satisfaction with the system. The system integrates tools from the Google Applications suite-namely Gmail, Google Docs, Google Drive, Google Forms, Google Sheets, and Google Slides-to enhance data management efficiency and facilitate system accessibility.System performance was evaluated by 15 experts across five dimensions: design, efficiency, usability, security, and resource utilization. The results indicated that the system achieved a 'very good' rating in four dimensions, with resource utilization rated as 'excellent'. Additionally, satisfaction was assessed among 13 users, including instructors and scientific staff, across six dimensions:design, system efficiency, usability, content, security, and resources. Users reported high satisfaction in system design and security, and the highest level of satisfaction in system efficiency, usability, content, and resource utilization. Based on the findings, the developed system effectively meets user needs and contributes to improving the quality of scientific equipment management for laboratory instruction.

Development of a Rule-Based Chatbot for Standardizing Undergraduate Research Project Report Preparation

Thipwan Fufaung — 2026-08-07

This research aims to develop and evaluate the performance of a chatbot utilizing a Rule-Based Approach to control the formatting standards of undergraduate research project reports. The research methodology comprised five stages: analysis, design, development, experimentation, and evaluation. The prototype was developed on the Chatfuel platform, with its operational mechanism described using the 3-Tier Architecture concept. The research instruments were validated for content validity (IOC) by two experts, yielding indices ranging from 0.50 to 1.00. The evaluation, involving 20 users, revealed that the chatbot received a good efficiency rating from experts (Mean = 4.00 on a 5-point scale), while user satisfaction was at a high level (Mean = 3.81, S.D. = 0.93). The content dimension received the lowest average score (3.80). Qualitative analysis indicated that the issues stemmed primarily from navigational obstacles rather than content quality. The results confirm that the Rule-Based Approach is sufficiently effective in controlling research report formatting quality. The study suggests enhancing the user experience to facilitate future development into a Hybrid Model.

Deep Learning and Kernel Density Estimation for Critical Area Identification: A Sustainable Tourism Management Case Study in Northern Thailand

Tobthong Chancharoen — 2026-08-07

Tourism in Northern Thailand is a strategic pillar of the regional economy, yet managing its sustainability remains a critical challenge. This research aims to analyze the factors driving dissatisfaction among international tourists toward natural attractions in Northern Thailand. A dataset of 8,193 reviews was collected from TripAdvisor, spanning a ten-year period from 2014 to 2023.

This study presents an academic contribution by integrating a BERT-based Deep Learning model to classify negative sentiments—the primary focus of this analysis. The model achieved a high precision of 87% in identifying negative class data. Subsequently, N-Gram Analysis was employed for issue clustering, and Kernel Density Estimation (KDE) with a bandwidth (h) of 0.3 km was applied to transform textual data into a continuous spatial density surface, enabling the precise identification of dissatisfaction hotspots.

The findings reveal that the primary driver of dissatisfaction is Value & Pricing, evidenced by the prevalence of the phrase 'not worth' (83 mentions) and specific complaints regarding high entry fees (e.g., '300 baht'). These issues are spatially correlated with environmental and maintenance concerns (e.g., 'no water' and 'dirty'). Negative hotspots are intensely concentrated in key locations, notably Doi Inthanon National Park and Pai District in Mae Hong Son Province.

The results provide actionable spatial insights to support government agencies in targeted policymaking. By identifying these crisis zones, the study facilitates informed decisions on revising entry fee structures and managing tourist carrying capacity within hotspots to enhance the long-term sustainability of Northern Thailand's natural tourism.

Development of Intelligent Motion-Detection and Image-Recording Software Using a Webcam for Application in Business Organization Security

Witthaya Boonsuk — 2026-08-07

The objectives of this research are: 1) to develop software applied for the surveillance of business organizations 2) to design a system that can automatically capture still images every10 seconds, facilitating convenient data storage and retrospective examination 3) to develop an automatic motion detection mechanism that records video only when movement is detected, continuous for 30 seconds before stopping, until new movement is detected. and 4) to test and compare the storage space efficiency of the developed software This approach helps reduce storage usage and enhances practical efficiency.

This research employs digital image processing techniques in combination with motion detection methods, namely frame differencing and background subtraction. The system is developed using C# and connected to a webcam to capture still images at specified intervals and record video only when motion is detected. All data are stored in digital file formats for convenient retrieval and review.

The evaluation results of the developed software, based on webcam recording tests, indicate that: Function 1(still image capture) achieved 88.15% storage savings.and Function 2 (motion-triggered video recording) achieved 98.75% storage savings. These results demonstrate that the software provides highly effective storage optimization and is suitable for application in monitoring and surveillance of business organizations as well as for further development and commercial.

Statistical Relationship Assessment and Temporal Variability Analysis of Water Turbidity Using Satellite Time-Series Data in the Kok River, Chiang Rai Province

Phummipat Oonban — 2026-08-07

This study aims to evaluate the statistical relationship between water turbidity and surface reflectance derived from satellite imagery and to analyze the temporal variability of turbidity in the Kok River using time-series Sentinel-2 satellite data acquired between 1 January 2023 to 31 March 2025, processed on the Google Earth Engine (GEE) platform. The rank correlation analysis between field turbidity (NTU) and NDSSI showed a strong inverse relationship, with Spearman’s rank correlation and Kendall’s rank correlation coefficients of –0.83 and –0.64, respectively, indicating an association between NDSSI values and the ordinal variation of turbidity in the Kok River. Temporal analysis revealed a seasonal pattern in turbidity, with low to moderate levels during the dry season and a rapid increase during the rainy season. The maximum turbidity observed in the dataset was 6,150 NTU, while the minimum NDSSI value detected was –0.52, occurring at different times. Anomaly analysis indicated that turbidity tended to increase during the rainy season. Turbidity values exceeding 720 NTU were classified as positive anomalies, whereas NDSSI values lower than –0.13 were classified as negative anomalies. However, anomaly detection for NDSSI was limited to periods with available satellite imagery, as some months during the main monsoon season did not meet the image selection criteria. Overall, the results suggest that time-series NDSSI derived from satellite imagery may be useful for examining turbidity variation trends and identifying anomalous turbidity conditions in the Kok River, Chiang Rai Province.

Assessment of Amphibian Diversity and Distribution in Doi Saket District, Chiang Mai Province, Thailand

Pitchaphon Athitphokinchok — 2026-08-07

This study aimed to assess the diversity, distribution, and species status of amphibians in selected areas of Doi Saket District, Chiang Mai Province. Field surveys and specimen collections were conducted at four forest sites near water sources and streams between June to August 2023, specifically during the nighttime. The results revealed a total of 183 amphibian specimens, comprising 6 families, 16 genera, and 23 species. The family Rhacophoridae exhibited the highest species richness with 7 species, the family Ranidae with 5 species. Limnonectes taylori was identified as the most abundant species and was distributed across all survey sites. Regarding biodiversity indices, Site 3 (Ban Pang Kamphaeng Hin) demonstrated the highest species richness with 14 species. Meanwhile, Site 1 (Ban Pang Nam Thu) and Site 3 recorded the highest Shannon-Wiener Diversity Index values (2.23), reflecting a moderately diverse ecosystem. Site 2 (Ban Pang Bong) showed the highest species evenness (0.88). Furthermore, according to the IUCN Red List assessment, all 23 species were categorized as Least Concern (LC). However, habitat loss driven by community expansion and land utilization, along with climate variability, remained critical threats directly correlated with the risk of population decline among amphibians in the area. These findings provide valuable baseline data for future planning, management, and sustainable conservation of local biological resources.

Participatory Solid Waste Management Using Technology and Learning Approaches to Promote a Circular Economy in Kanchanaburi Municipality, Thailand

Narongsak Theanyai — 2026-08-07

This research aimed to: (1) examine the municipal solid waste management situation in Kanchanaburi Municipality; (2) develop a mobile application to support solid waste management for households within Kanchanaburi Municipality; and (3) investigate the potential for community-level value creation from municipal solid waste. A mixed-methods research design was employed, including data collection through questionnaires, interviews, focus group discussions, and participatory workshops. The findings revealed that most residents demonstrated positive attitudes toward waste management and awareness of waste-related problems; however, concrete community participation remained limited. The integration of technology and learning activities was found to enhance understanding and promote appropriate waste separation behaviors suited to an urban community context. The researchers developed a waste management application designed to meet user needs, incorporating key functions such as waste collection day notifications, waste classification information, and learning promotion activities. The evaluation results based on the Technology Acceptance Model (TAM) and user satisfaction indicated the highest level of acceptance and satisfaction. In addition, the research team developed value-added waste innovation initiatives, all of which were evaluated as suitable for urban communities facing spatial limitations and seeking environmentally friendly waste management solutions. Project participants were able to practically apply the acquired knowledge, disseminate it within the community, and generate income, with overall outcomes rated at a high level.

Development of a Sugarcane Plantation Classification Model Using Sentinel-1 and Sentinel-2 Satellite Imagery

Arket Sooksomnuek — 2026-08-07

Phu Khiao District, Chaiyaphum Province, is a significant sugarcane-growing area in northeastern Thailand, both in terms of cultivation area and production volume serving as the primary raw material for the sugar industry. However, sugarcane area mapping still relies largely on field surveys, leading to limitations in spatial detail and temporal consistency. This study develops a sugarcane plantation classification model using a Random Forest algorithm with 21 features comprising 13 Sentinel-2 spectral bands, 6 vegetation indices, and VV and VH polarizations from Sentinel-1, processed on Google Earth Engine. A total of 400 training polygons (54,304 pixels) and 400 independent field validation points were used. Four feature datasets were compared: (1) RGB optical data, (2) vegetation indices (VIs), (3) vegetation indices combined with SAR data (VIs + SAR), and (4) SAR data alone. The results indicate that integrating vegetation indices with SAR data (VIs + SAR) yields the best performance, achieving an Overall Accuracy of 88.50%, a Kappa Coefficient of 81.90%, and a Producer's Accuracy for sugarcane of 94.0%. The Red-edge indices, particularly CI red edge and SAREI, effectively reduced spectral overlap between sugarcane and other vegetation types, as sugarcane prior to harvest exhibits high and uniform chlorophyll content, resulting in a narrow interquartile range that is clearly distinguishable from other crops. These findings suggest that multi-sensor integration with Red-edge indices is effective for sugarcane plantation classification and can serve as a prototype for large-scale sugarcane monitoring.

Longevity Tourism: A Systematic Literature Synthesis and Research Directions in the Thai Context

Warach Madhyamapurush — 2026-08-07

Longevity Tourism is a rapidly growing research domain driven by global population ageing and heightened health consciousness. This article aims to (1) systematically synthesize the literature on longevity tourism; (2) identify research gaps in the Southeast Asian context, particularly Thailand; and (3) propose an integrative conceptual framework and future research directions. The authors employed a Systematic Literature Synthesis (SLS) protocol, searching Scopus, Web of Science, PubMed, and Google Scholar using pre-specified keyword strings, and screening results against PICOS criteria. A total of 25 works were included from 889 initially identified records. Synthesized findings were organized into five thematic clusters: definitional and theoretical foundations; Blue Zone tourism; health impacts on older tourists; integrative medicine and tourism; and Southeast Asian contexts. Asian scholarship predominantly situates longevity tourism within recognized longevity regions, while Western scholarship foregrounds service design and policy for ageing societies. Critical gaps identified include a near-absence of empirical work in the Thai context, a dearth of longitudinal research designs, and the lack of a domain-specific theoretical framework for Longevity Tourism. This article proposes the Integrative Longevity Tourism Framework (ILTF) as a foundation for future empirical inquiry in Thailand.