Figure 3 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Examples of photographic images over a range of brightness levels below the threshold used in LandSense. Brightness level of images shown (from left to right) are 2,8,36,46,56,68,81 and 94.
Source publication
The potential of citizens as a source of geographical information has been recognized for many years. Such activity has grown recently due to the proliferation of inexpensive location aware devices and an ability to share data over the internet. Recently, a series of major projects, often cast as citizen observatories, have helped explore and devel...
Context in source publication
Context 1
... the proportion of photographs failing the brightness check varied greatly in the pilots with up to half of the acquired images acquired in a pilot study being viewed as too dark (Table 2). Figure 3 illustrates the quality of images that fail to reach the threshold of 100 used by LandSense. It is evident that images at the lower end of the spectrum convey little visual information. ...
Citations
... But now, an exciting new approach is changing the way we collect data, thanks to the power of crowdsourcing. A team of enthusiastic volunteers equipped with GPS devices and or online visualization platforms all contributed to the collection of land cover information (Foody et al. 2022;Fraisl et al. 2022;Tavra, Racetin, and Peroš 2021;Zhao et al. 2017). Whereas, each method of sample collection has its disadvantages: (1) collecting samples through field surveys is the most resource-consuming; ...
High-quality samples for training and validation are crucial for land cover classification, especially in some complex scenarios. The reliability, representativeness, and generalizability of the sample set are important for further researches. However, manual interpretation is subjective and prone to errors. Therefore, this study investigated the following questions: (1) How much difference is there in the interpreters’ performance across educational levels? (2) Do the accuracies of human and AI (Artificial Intelligence) improve with increased training and supporting material? (3) How sensitive are the accuracies of land cover types to different supporting material? (4) Does interpretation accuracy change with interpreters’ consistency? The experiment involved 50 interpreters completing five cycles of manual image interpretation. Higher educational background interpreters showed better performance: accuracies pre-training at 52.22% and 58.61%, post-training at 61.13% and 70.21%. Accuracy generally increased with more supporting material. Ultra-high-resolution images and background knowledge contributed the most to accuracy improvement, while the time series of normalized difference vegetation index (NDVI) contributed the least. Group consistency was a reliable indicator of volunteer sample reliability. In the case of limited samples, AI was not as good as manual interpretation. To ensure quality in samples through manual interpretation, we recommend inviting educated volunteers, providing training, preparing effective support material, and filtering based on group consistency.
... Errors in LUC categorization are rather typical. In order to assess the accuracy of the mapping, the classification in this study involved the construction of a matrix error, which is a common way to show the accuracy of classification [13,30], sampling description, and proportion of each category in the map. The overall accuracy, producer and user accuracies, and kappa coefficient were determined from the created error matrix. ...
Hippopotamus ( Hippopotamus amphibius ) are victims of land use changes (LUC) due to their semi-aquatic nature. Although human–hippopotamus interactions (HHI) are known to exist in the area surrounding Lake Babati, nothing is known about the LUC in relation to the interactions. The study aimed at assessing the trend of LUC in the last 20 years in relation to time of establishing new settlement and farming seasons in relation to HHI, respondents’ perceptions of HHI, and mitigation measures used by local people against Hippos adjacent Lake Babati. Remote sensing and GIS techniques, questionnaires and focused group discussions were used to assess human perceptions regarding trends of the LUC in the study area. LUC was monitored by using landsat images from the years 1999 and 2019. The findings indicate an increase in settlement while water, agroforestry, and seasonal agricultural lands were decreasing. The time respondents stayed in the village, farm size, and respondents’ perception of HHI trends were observed to vary with the distance from the lake. The presence of LUC on adjacent Lake Babati jeopardized the ecological integrity of Hippos’ habitat and increased tension and overlap between hippos’ and human needs. The findings provide a baseline for managing HHI and recommend proper land use planning that prioritizes the use of alternative crops like fruit trees especially within 3 km from the lake. Human population and settlement expansion patterns should also be monitored in areas closer to the lake for sustaining wildlife conservation and livelihood development in Lake Babati and surrounding areas.
