Using deep learning semantic segmentation for land use extraction is the most challenging problem in medium spatial resolution imagery. This is because of the deep convolution layer and multiple levels of deep steps of the baseline network, which can cause a degradation problem in small land use features. In this paper, a deep learning semantic segmentation algorithm which comprises an adjustment network architecture (LoopNet) and land use dataset is proposed for automatic land use classification using Landsat 8 imagery. The experimental results illustrate that deep learning semantic segmentation using the baseline network (SegNet, U-Net) outperforms pixel-based machine learning algorithms (MLE, SVM, RF) for land use classification. Furthermore, the LoopNet network, which comprises a convolutional loop and convolutional block, is superior to other baseline networks (SegNet, U-Net, PSPnet) and improvement networks (ResU-Net, DeeplabV3+, U-Net++), with 89.84% overall accuracy and good segmentation results. The evaluation of multispectral bands in the land use dataset demonstrates that Band 5 has good performance in terms of extraction accuracy, with 83.91% overall accuracy. Furthermore, the combination of different spectral bands (Band 1–Band 7) achieved the highest accuracy result (89.84%) compared to individual bands. These results indicate the effectiveness of LoopNet and multispectral bands for land use classification using Landsat 8 imagery.

Coastal erosion is a critical environmental challenge in the Upper Gulf of Thailand, driven by both natural processes and human activities. This study analyzes 35 years (1988–2023) of shoreline changes using geoinformatics, machine learning algorithms (Random Forest, Support Vector Machine, Maximum Likelihood, Minimum Distance), and the Digital Shoreline Analysis System (DSAS). The results show that the Random Forest algorithm, utilizing spectral bands and indices (NDVI, NDWI, MNDWI, SAVI), achieved the highest classification accuracy (98.17%) and a Kappa coefficient of 0.9432, enabling reliable delineation of land and water boundaries. The extracted annual shorelines were validated with high accuracy, yielding RMSE values of 13.59 m (2018) and 8.90 m (2023). The DSAS analysis identified significant spatial and temporal variations in shoreline erosion and accretion. Between 1988 and 2006, the most intense erosion occurred in regions 4 and 5, influenced by sea-level rise, strong monsoonal currents, and human activities. However, from 2006 to 2018, erosion rates declined significantly, attributed to coastal protection structures and mangrove restoration. The period 2018–2023 exhibited a combination of erosion and accretion, reflecting dynamic sediment transport processes and the impact of coastal management measures. Over time, erosion rates declined due to the implementation of protective structures (e.g., bamboo fences, rock revetments) and the natural expansion of mangrove forests. However, localized erosion remains persistent in low-lying, vulnerable areas, exacerbated by tidal forces, rising sea levels, and seasonal monsoons. Anthropogenic activities, including urban development, mangrove deforestation, and aquaculture expansion, continue to destabilize shorelines. The findings underscore the importance of sustainable coastal management strategies, such as mangrove restoration, soft engineering coastal protection, and integrated land-use planning. This study demonstrates the effectiveness of combining machine learning and geoinformatics for shoreline monitoring and provides valuable insights for coastal erosion mitigation and enhancing coastal resilience in the Upper Gulf of Thailand.

Urban building segmentation from remote sensed imageries is challenging due to there usually existing a variety of building features. Furthermore, very high spatial resolution imagery can provide many details of the urban building, such as styles, small gaps among buildings, building shadows, etc. Hence, satisfactory accuracy in detecting and extracting urban features from highly detailed images still remains. Deep learning semantic segmentation using baseline networks works well on building extraction, however their ability in building extraction in shadows area, unclear building feature, and narrow gaps among buildings in dense building zone is still limited. In this paper, we propose parallel cross learning based pixel transferred deconvolutional network (PCL-PTD net), and then is used to segment urban buildings from aerial photographs. The proposed method is evaluated and intercompared with traditional baseline networks. In PCL-PTD net, it is composed of parallel network, cross learning functions, residual unit in encoder part, and pixel transferred deconvolution in decoder part. The performance is applied to three datasets (Inria aerial dataset, ISPRS Potsdam dataset, UAV building dataset), to evaluate its accuracy and robustness. As a result, we found that PCL-PTD net can improve learning capacities of the supervised learning model in differentiating buildings in dense area and extracting buildings covered by shadows. As compared to the baseline networks, we found that proposed network shows superior performance compared to all eight networks (SegNet, U-net, PSPnet, PixelDCL, DeeplabV3+, U-Net++, CFENet, IRU-Net). The experiments on three datasets also demonstrate the ability of proposed framework and indicating its performance.

Rice extraction is critical in remote sensing, especially in Suphan Buri province, Thailand, using Sentinel-1 synthetic aperture radar (SAR) time-series data and advanced machine learning algorithms. Given the challenges of varied terrains and diverse crop types, the research employs different polarization modes (vertical transmit and vertical receive (VV), vertical transmit and horizontal receive (VH), and VV+VH) to enhance classification accuracy. The study evaluates the performance of three machine learning algorithms: random forest, extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM). The results demonstrate that combined VV+VH polarization outperforms VV and VH alone, providing better accuracy due to its ability to capture more detailed object features. LightGBM emerged as the most effective among the algorithms, particularly when dealing with large datasets. After hyperparameter tuning (n_estimators: 820, max_depth: 10, and learning_rate: 0.01), LightGBM achieved the highest accuracy. The rice class showed exceptional precision, recall, and F1-score, surpassing other land-use classes (agriculture/forest and urban areas). However, these classes still pose challenges, highlighting the need for future studies to integrate multi-sensor data and explore more sophisticated machine-learning models. This research offers a promising approach to enhancing rice monitoring and management in diverse agricultural landscapes, contributing to more accurate and efficient farming practices.

This study presents the distribution of road accidents in Bangkok, Thailand, during the COVID-19 outbreak (162 days) using spatial statistics. The findings show that the number of road accidents fell by 28.53% during the lockdown period (March 26 to June 14, 2020) compared to before the lockdown period (January 5 to March 25, 2020). There was also a considerable decrease in road accidents by approximately 4.96% compared to the same period in 2019 (January 4 to June 14, 2019). High accident incidence areas were located in central and southwestern Bangkok, and this pattern was also observed before and during the lockdown as well as in 2019. The findings revealed the numbers and patterns of road accidents that occurred during the COVID-19 outbreak. This information can be useful in minimizing and limiting the impact and damage caused by road accidents in Bangkok. Keywords: COVID-19; lockdown; road accident; Moran’s I; spatial autocorrelation *Corresponding author: Tel.: (+66) 822313386 Fax: (+66) 38102379 E-mail: narong_p@buu.ac.th

A water consumption assessment using a bottom-up approach is applied in this research; the smallest level of water consumption assessment shows the water consumption behavior at the farm level of 55 mangosteen farms in the Khlung District of Chanthaburi Province, Thailand, in the production year 2019. The findings revealed that the average water footprint (WF) was 774.60 m3/ton, which was divided into a green water footprint of 519.04 m3/ton and the blue water footprint of 255.56 m3/ton. Stage 5 was the mangosteen’s growth stage that had the highest WF: this stage was the fruit’s maturation period, whose WF was equivalent to 41.16% of the yearly water consumption. The WF of mangosteen data for the production year 2019 is a crucial baseline that will enable farmers to understand the actual water consumption in mangosteen production at the farm level. It will be feasible to determine the trend of changing water use, particularly if the mangosteen WF data is regularly gathered each year and it has led to appropriate water consumption planning per the needs of the mangosteen. Furthermore, this research also raised farmers’ awareness concerning the water consumption of mangosteen production.