Monitoring Coastline Changes of the Malay Islands Based on Google Earth Engine and Dense Time-Series Remote Sensing Images

The use of remote sensing to monitor coastlines with wide distributions and dynamic changes is significant for coastal environmental monitoring and resource management. However, most current remote sensing information extraction of coastlines is based on the instantaneous waterline, which is obtained by single-period imagery. The lack of a unified standard is not conducive to the dynamic change monitoring of a changeable coastline. The tidal range observation correction method can be used to correct coastline observation to a unified climax line, but it is difficult to apply on a large scale because of the distribution of observation sites. Therefore, we proposed a coastline extraction method based on the remote sensing big data platform Google Earth Engine and dense time-series remote sensing images. Through the instantaneous coastline probability calculation system, the coastline information could be extracted without the tidal range observation data to achieve a unified tide level standard. We took the Malay Islands as the experimental area and analyzed the consistency between the extraction results and the existing high-precision coastline thematic products of the same period to achieve authenticity verification. Our results showed that the coastline data deviated 10 m in proportion to a reach of 40% and deviated 50 m within a reach of 89%. The overall accuracy was kept within 100 m. In addition, we extracted 96 additional islands that have not been included in public data. The obtained multi-phase coastlines showed the spatial distribution of the changing hot regions of the Malay Islands’ coastline, which greatly supported our analysis of the reasons for the expansion and retreat of the coastline in this region. These research results showed that the big data platform and intensive time-series method have considerable potential in large-scale monitoring of coastline dynamic change and island reef change monitoring.

Hybridizing Cross-Level Contextual and Attentive Representations for Remote Sensing Imagery Semantic Segmentation

Semantic segmentation of remote sensing imagery is a fundamental task in intelligent interpretation. Since deep convolutional neural networks (DCNNs) performed considerable insight in learning implicit representations from data, numerous works in recent years have transferred the DCNN-based model to remote sensing data analysis. However, the wide-range observation areas, complex and diverse objects and illumination and imaging angle influence the pixels easily confused, leading to undesirable results. Therefore, a remote sensing imagery semantic segmentation neural network, named HCANet, is proposed to generate representative and discriminative representations for dense predictions. HCANet hybridizes cross-level contextual and attentive representations to emphasize the distinguishability of learned features. First of all, a cross-level contextual representation module (CCRM) is devised to exploit and harness the superpixel contextual information. Moreover, a hybrid representation enhancement module (HREM) is designed to fuse cross-level contextual and self-attentive representations flexibly. Furthermore, the decoder incorporates DUpsampling operation to boost the efficiency losslessly. The extensive experiments are implemented on the Vaihingen and Potsdam benchmarks. In addition, the results indicate that HCANet achieves excellent performance on overall accuracy and mean intersection over union. In addition, the ablation study further verifies the superiority of CCRM.

2-DIMENSIONAL GEOMETRIC ANALYSIS OF A SIMPLE FREE NETWORK

This paper attempts to quantify geometric considerations in observations and observe trends in solutions to free network solutions. The method of investigation will be utilizing 2D observations to determine how each measurement affects the overall solution and the location of the observations relative to the other nodes. A local reference system will be determined using the Gauss-Markov model with constraints by fixing the largest range observation to the y-axis to give a relative orientation. Further solutions will be calculated by fixing additional points to generate multiple least squares solutions relative to the local reference system. The resulting final points will be modeled using the Gauss-mixture model and compared to a simulated dataset generated by adding random error to the observations. Different weight matrices will be tested to demonstrate the effect on the overall solution. These methods were chosen because of prior experimentation by different research groups studying geometric considerations for UAS and ground surveying conditions. The major contribution will be the trends observed in the modeling and the correlation of the fixed local solutions to the geometry of the points.

GPS Receiver Position Interpretation using Single Point PVT Estimation Algorithm

In our day-to-day lives, we need to get the correct GPS location information. GPS is based on the calculation of the pseudo-range and four unspecified parameters, but the formula is not linear in navigation observation. A single point position algorithm can solve the nonlinear equation; the algorithm is based on Taylor linearization. This paper provides an overview of the single point PVT algorithm and presents the GPS satellite pseudo-range observation equations, typically over-determined as there are only four unknown satellites, but generally, more than four are monitored and thus more than four pseudo-range observation equations. Single point PVT estimation algorithm is used to solve pseudo range observation equations in order to find position and clock bias solutions are described in detail. In this article, the position of GPS receiver is estimated w.r.t. to X, Y, Z Coordinates, in addition to that clock bias also estimated.