Research on fine tuning parameter insulator identification based on YOLOV4 algorithm

Aiming at the problem that the training network time of YOLOV4 algorithm is too long due to the large data set of aerial insulator images, a method based on YOLOV4 algorithm is proposed to shorten the training time by fine-tuning parameters without affecting the positioning detection accuracy. Based on the development of UBANTU virtual machine, through CUDA and CUDNN environment configuration, and through the detection and verification of insulator aerial photo data set, the feasibility of accurate positioning of insulators under the condition of fine tuning parameters of YOLOV4 algorithm is successfully proved.

Utilization of Unmanned Aerial Vehicle (UAV) Technology for Mapping Mangrove Ecosystem

Technological developments, especially in remote sensing in mangrove mapping are growing. One of them is the use of Unmanned Aerial Vehicle (UAV) as a vehicle for capturing aerial photo data. This study aims to map the mangrove ecosystem in order to find out spatial information with UAV technology and to identify mangrove species, distribution, and associations. The ground survey was conducted on 02 until 04 May 2019 in North of Lancang Island, Seribu Islands, DKI Jakarta Province. The mangrove data collection was carried out in four stations with marking, tracking the mangrove area and drone flights for taking aerial photo data. Based on the research, it was found that the mangroves were in good condition with an area of around 4 hectares, consisting of Rhizophora mucronata. However, there is a lot of waste around the mangrove ecosystem, especially inorganic waste in the form of plastic. Where the presence of garbage can cover the roots of the mangrove so that it can affect respiration and cause death in mangroves. The application of UAV technology that is integrated with GIS in mangrove ecosystem mapping is expected to be an alternative in extracting mangrove databases for future coastal ecosystem management.

The accuracy of PlanetScope imagery to nitrogen, phosphorus, potassium and sulfur nutrition estimation in terraced paddy field

This study aims to find out the accuracy of PlanetScope imagery in analyzing the total content of Nitrogen (N), Phosphorus (P), Potassium (K) and Sulfur (S) in terrace paddy field. The area for this research is terraced paddy fields in Wonosari Sub-district, Gondangrejo District, Karanganyar Regency, Central Java Province, Indonesia. Aerial photo analysis from PlanetScope (www.planet.com) was conducted at the Pedology Laboratory and Soil Survey, while soil sample analysis was carried out at the Laboratory of Chemistry and Soil Fertility, Faculty of Agriculture, Sebelas Maret University (UNS). The preparation aerial photo analysis from PlanetScope was carried out in September 2019. Soil sampling, laboratory analysis, and data analysis were carried out from September 2019 to January 2020. There are 3 Citra PlanetScope bands that have a significant effect on the elements being studied. Those three bands are band 1 (red), band 3 (blue) and band 4 (NIR). Multiple linear regression analysis obtained a regression equation model from those three bands. This model can be used as an estimator for the existence of the soil nutrient being studied. The accuracy values obtained for N, P, K and S in terraced paddy fields are 97.58%; 94.85%; 85.97%; and 97.39% and the R2 value of 0.39; 0.30; 0.30 and 0.27. The R2 value, which is still relatively small, indicates that there are many factors may affect the accuracy value. Int. J. Agril. Res. Innov. Tech. 11(1): 1-9, June 2021

Multi-scale remote sensing observations of a palsa in degradation phase

<p>The Vissátvuopmi palsa complex (N 68°74′50′′, E 21°11′30”) is the largest coherent palsa complex in Sweden (ca 274 ha). Aerial photo-interpretation over an area covered by plateau palsas showed a 30% decline in lateral area -- from ca 70 to 49 ha -- that occurred between 1955 to 2016 (Olvmo et al., 2020). Within Vissátvuopmi, we have more closely studied two single palsas, one dome-shaped and one ridge-shaped, for changes in extent, height and vegetation composition. Manual interpretation of aerial photography between 1955 and 2016 show lateral degradation of 35% and 54% for the dome and ridge palsas, respectively. Since 2018 we have monitored the palsas using images from drones as well as analysis of Planet Dove and Sentinel-2 satellite imagery. Photogrammetry is used to produce orthophotos as well as digital surface models (DSMs) from the drone images, and compared to earlier LiDAR and aerial photo DSMs, to study lateral and vertical degradation.</p><p>The drone-generated DSMs from 2018, 2019 and 2020 show further lateral degradation of the two large palsas. In 2020 a rapid change in vegetation composition was seen on the dome-shaped palsa, where a 250 m<sup>2</sup> area of <em>Betula nana</em> and <em>Empetrum hermaphroditum</em> transitioned to lichen. This vegetation change could be seen in spectral data from both drone and satellite platforms. The future development of this palsa, monitored annually using both fine and medium spatial resolution data, will give insight into the timing and signs of the individual palsas in stages of degradation.</p>

Clarketal_2021.zip

This dataset is supplementary to a submitted manuscript to MDPI: Geosciences, "Making an Impact" special issue.<div><br></div><div>Manuscript title: Post-impact faulting of the Holfontein Granophyre Dike of the Vredefort Impact Structure, South Africa, inferred from remote sensing, geophysics, and geochemistry.</div><div>Authors: Clark, M.D., Kovaleva, E., Huber, M.S., Fourie, F., Harris, C.<br><div><br></div><div>The file contains an aerial photo and videos acquired via a drone which document the spatial expression of the geological dike.</div></div><div><br></div><div>Resolution of video is 4k (3840x2160 pix). Resolution of still images is 5472x3648 pix.</div>

Clarketal_2021.zip

This dataset is supplementary to a submitted manuscript to MDPI: Geosciences, "Making an Impact" special issue.<div><br></div><div>Manuscript title: Post-impact faulting of the Holfontein Granophyre Dike of the Vredefort Impact Structure, South Africa, inferred from remote sensing, geophysics, and geochemistry.</div><div>Authors: Clark, M.D., Kovaleva, E., Huber, M.S., Fourie, F., Harris, C.<br><div><br></div><div>The file contains an aerial photo and videos acquired via a drone which document the spatial expression of the geological dike.</div></div><div><br></div><div>Resolution of video is 4k (3840x2160 pix). Resolution of still images is 5472x3648 pix.</div>

Clarketal_2021.zip

This dataset is supplementary to a submitted manuscript to MDPI: Geosciences, "Making an Impact" special issue.<div><br></div><div>Manuscript title: Post-impact faulting of the Holfontein Granophyre Dike of the Vredefort Impact Structure, South Africa, inferred from remote sensing, geophysics, and geochemistry.</div><div>Authors: Clark, M.D., Kovaleva, E., Huber, M.S., Fourie, F., Harris, C.<br><div><br></div><div>The file contains an aerial photo and videos acquired via a drone which document the spatial expression of the geological dike.</div></div><div><br></div><div>Resolution of video is 4k (3840x2160 pix). Resolution of still images is 5472x3648 pix.</div>