As a convenience to ASPRS members, in-press peer reviewed articles approved for publication in forthcoming issues of PE&RS are listed below.

Manuscript number: 22-00132R3
Land Use Change in the Yangtze River Economic Belt during 2010 to 2020 and Future Comprehensive Prediction Based on Markov and ARIMA Models
Haotian Zheng, Fan Yu, Huawei Wan, Peirong Shi, and Haonan Wang

The key data for accurate prediction is of great significance to accurately carry out the next step of sustainable land use development plan according to the demand of China. Conse-quently, the main purposes of our study are : (1) to delineate the characteristics of land use transitions within the Yangtze River Economic Belt; (2) to use the Markov model and the autoregressive integrated moving average (ARIMA) model for comparative analysis and prediction of land use distribution. This study analyzes land use/cover change (LUCC) data from 2010 and 2020 using the land use transition matrix, dynamic degree, and compre-hensive index model and predicts 2025 land use by the Markov model.. The study identifies a reduction in land usage over 11 years, particularly in grassland. The Markov and ARIMA models’ significance is 0.002 (P < 0.01) , showing arable land and woodland dominance, with varying changes in other land types.

Manuscript number: 23-00074R2
An Improved YOLO Network for Insulator and Insulator Defect Detection in UAV Images
Fangrong Zhou, Lifeng Liu, Hao Hu, Weishi Jin, Zezhong Zheng, Zhongnian Li, Yi Ma, and Qun Wang

The power grid plays a vital role in the construction of livelihood projects by transmitting electrical energy. In the event of insulator explosions on power grid towers, these insulators may detach, presenting potential safety risks to transmission lines. The identification of such failures relies on the examination of images captured by unmanned aerial vehicles (UAVs). However, accurately detecting insulator defects remains challenging, particularly when dealing with variations in size. Existing methods exhibit limited accuracy in detecting small objects. In this paper, we propose a novel detection method that incorporates the convolutional block attention module (CBAM) as an attention mechanism into the backbone of the “you only look once” version 5 (YOLOv5) model. Additionally, we integrate a residual structure into the model to learn additional information and features related to insulators, thereby enhancing detection efficiency. Experimental results demonstrate that our proposed method achieved F1 scores of 0.87 for insulator detection and 0.89 for insulator defect detection. The improved YOLOv5 network shows promise in detecting insulators and their defects in UAV images.

Manuscript number: 23-00083R2
Real-Time Semantic Segmentation of Remote Sensing Images for Land Management
Yinsheng Zhang, Ru Ji, Yuxiang Hu, Yulong Yang, Xin Chen, Xiuxian Duan, and Huilin Shan

Remote sensing image segmentation is a crucial technique in the field of land management. However, existing semantic segmentation networks require a large number of floating-point operations (FLOPs) and have long run times. In this paper, we propose a dual-path feature aggregation network (DPFANet) specifically designed for the low-latency operations required in land management applications. Firstly, we use four sets of spatially separable convolutions with varying dilation rates to extract spatial features. Additionally, we use an improved version of MobileNetV2 to extract semantic features. Furthermore, we use an asymmetric multi-scale fusion module and dual-path feature aggregation module to enhance feature extraction and fusion. Finally, a decoder is constructed to enable progressive up-sampling. Experimental results on the Potsdam data set and the Gaofen image data set (GID) demonstrate that DPFANet achieves overall accuracy of 92.2% and 89.3%, respectively. The FLOPs are 6.72 giga and the number of parameters is 2.067 million.

Manuscript number: 24-00001R2
Monitoring an Ecosystem in Crisis: Measuring Seagrass Meadow Loss Using Deep Learning in Mosquito Lagoon, Florida
Stephanie A. Insalaco, Hannah V. Herrero, Russ Limber, Clancy Oliver, and William B. Wolfson

The ecosystem of Mosquito Lagoon, Florida, has been rapidly deteriorating since the 2010s, with a notable decline in keystone seagrass species. Seagrass is vital for many species in the lagoon, but nutrient overloading, algal blooms, boating, manatee grazing, and other factors have led to its loss. To understand this decline, a deep neural network analyzed Landsat imagery from 2000 to 2020. Results showed significant seagrass loss post-2013, coinciding with the 2011–2013 super algal bloom. Seagrass abundance varied annually, with the model performing best in years with higher seagrass coverage. While the deep learning method successfully identified seagrass, it also revealed that recent seagrass coverage is almost non-existent. This monitoring approach could aid in ecosystem recovery if coupled with appropriate policies for Mosquito Lagoon's restoration.

Manuscript number: 24-00003R2
ReLAP-Net: Residual Learning and Attention Based Parallel Network for Hyperspectral and Multispectral Image Fusion
Aditya Agrawal, Souraja Kundu, Touseef Ahmad, Manish Bhatt

Remote sensing applications require high-resolution images to obtain precise information about the Earth’s surface. Multispectral images have high spatial resolution but low spectral resolution. Hyperspectral images have high spectral resolution but low spatial resolution. This study proposes a residual learning and attention-based parallel network based on residual network and channel attention. The network performs image fusion of a high spatial resolution multispectral image and a low spatial resolution hyperspectral image. The network training and fusion experiments are conducted on four public benchmark data sets to show the effectiveness of the proposed model. The fusion performance is compared with classical signal processing–based image fusion techniques. Four image metrics are used for the quantitative evaluation of the fused images. The proposed network improved fusion ability by reducing the root mean square error and relative dimensionless global error in synthesis and increased the peak signal-to-noise ratio when compared to other state-of-the-art models .

Manuscript number: 24-00016R1
Assessing the Utility of Uncrewed Aerial System Photogrammetrically Derived Point Clouds for Land Cover Classification in the Alaska North Slope
Jung-kuan Liu, Rongjun Qin, and Samantha T. Arundel

Uncrewed aerial systems (UASs) have been used to collect “pseudo field plot” data in the form of large-scale stereo imagery to supplement and bolster direct field observations to monitor areas in Alaska. These data supplement field data that is difficult to collect in such a vast landscape with a relatively short field season. Dense photogrammetrically derived point clouds are created and are facilitated to extract land cover data using a support vector machine (SVM) classifier in this study. We test our approach using point clouds derived from 1-cm stereo imagery of plots in the Alaska North Slope region and compare the results to field observations. The results show that the overall accuracy of six land cover classes (bare soil, shrub, grass, forb/herb, rock, and litter) is 96.8% from classified patches. Shrub had the highest accuracy (>99%) and forb/herb achieved the lowest (<48%). This study reveals that the approach could be used as reference data to check field observations in remote areas.

Manuscript number: 23-00070R2
Enhancing Forest–Steppe Ecotone Mapping Accuracy through Synthetic Aperture Radar–Optical Remote Sensing Data Fusion and Object-based Analysis
Ruilin Wang, Meng Wang, Xiaofang Sun, Junbang Wang, and Guicai Li

In ecologically vulnerable regions with intricate land use dynamics, such as ecotones, frequent and intense land use transitions unfold. Therefore, the precise and timely mapping of land use becomes imperative. With that goal, by using principal component analysis, we integrated Sentinel-1 and Sentinel-2 data, using an object-oriented methodology to craft a 10-meter-resolution land use map for the forest–grassland ecological zone of the Greater Khingan Mountains spanning the years 2019 to 2021. Our research reveals a substantial enhancement in classification accuracy achieved through the integration of synthetic aperture radar–optical remote sensing data. Notably, our products outperformed other land use/land cover data sets, excelling particularly in delineating intricate riverine wetlands. The 10-meter land use product stands as a pivotal guide, offering indispensable support for sustainable development, ecological assessment, and conservation endeavors in the Greater Khingan Mountains region. 

Manuscript number: 23-00085R2
Dynamic Monitoring of Ecological Quality in Eastern Ukraine Amidst the Russia-Ukraine Conflict
Chaofei Zhang, Zhanghua Xu, Yuanyao Yang, Lei Sun, and Haitao Li

To evaluate the spatiotemporal changes in the ecological environment of eastern Ukraine since the Russia-Ukraine conflict, this study used MODIS images from March to September 2020 and 2022 to calculate the Remote Sensing–Based Ecological Index. In 2022, compared with 2020, conflict zones exhibited reduced improvement and increased slight degradation, whereas nonconflict areas showed marginal enhancement. Through propensity score matching, the research confirmed the causal relationship between conflict and ecological trends. Pathway analysis revealed that the conflict contributed to 0.016 units increase in ecological quality while reducing the improvement rate by 0.042 units. This study provides empirical support for understanding the correlation between conflicts and specific environmental factors, offering technical references for ecological quality assessments in other conflict areas and future evaluations by the Ukrainian government.

Manuscript number: 24-00013R2
A Surface Water Extraction Method Integrating Spectral and Temporal Characteristics
Yebin Zou and Rui Shu

Remote sensing has been applied to observe large areas of surface water to obtain higher-resolution and long-term continuous observation records of surface water. However, limitations remain in the detection of large-scale and multi-temporal surface water mainly due to the high variability in water surface signatures in space and time. In this study, we developed a surface water remote sensing information extraction model that integrates spectral and temporal characteristics to extract surface water from multi-dimensional data of long-term Landsat scenes to explore the spatiotemporal changes in surface water over decades. The goal is to extract open water in vegetation, clouds, terrain shadows, and other land cover backgrounds from medium-resolution remote sensing images. The average overall accuracy and average kappa coefficient of the classification were verified to be 0.91 and 0.81, respectively. Experiments applied to China's inland arid area have shown that the method is effective under complex surface environmental conditions.

Manuscript number: 24-00008R2
Development of an Automatic Feature Point Classification Method for Three-Dimensional Mapping Around Slewing and Derricking Cranes
Hisakazu Shigemori, Junichi Susaki, Mizuki Yoneda, and Marek Ososinski

Crane automation requires a three-dimensional (3D) map around cranes that should be reconstructed and updated quickly. In this study, a high-precision classification method was developed to distinguish stationary objects from moving objects in moving images captured by a monocular camera to stabilize 3D reconstruction. To develop the method, a moving image was captured while the crane was slewed with a monocular camera mounted vertically downward at the tip of the crane. The boom length and angle data were output from a control device, a controller area network. For efficient development, a simulator that imitated the environment of an actual machine was developed and used. The proposed method uses optical flow to track feature points. The classification was performed successfully, independent of derricking motion. Consequently, the proposed method contributes to stable 3D mapping around cranes in construction sites.

Manuscript number: 23-00076R2
Semantic Segmentation of Point Cloud Scene via Multi-Scale Feature Aggregation and Adaptive Fusion
Baoyun Guo, Xiaokai Sun, Cailin Li, Na Sun, Yue Wang, and Yukai Yao

Point cloud semantic segmentation is a key step in 3D scene understanding and analysis. In recent years, deep learning–based point cloud semantic segmentation methods have received extensive attention from researchers. Multi-scale neighborhood feature learning methods are suitable for inhomogeneous density point clouds, but different scale branching feature learning increases the computational complexity and makes it difficult to accurately fuse different scale features to express local information. In this study, a point cloud semantic segmentation network based on RandLA-Net with multi-scale local feature aggregation and adaptive fusion is proposed. The designed structure can reduce computational complexity and accurately express local features. The mean intersection-over-union is improved by 1.1% on the SemanticKITTI data set with an inference speed of nine frames per second, while the mean intersection-over-union is improved by 0.9% on the S3DIS data set, compared with RandLA-Net. We also conduct ablation studies to validate the effectiveness of the proposed key structure.