Digitize land use (two classes: 1. Urban-Dense, 2. Settlement-Agriculture) in the Dosso region of Niger.
Urban-dense areas contain a cluster of buildings/huts that exceed 12 or so...or could be 1-2 large commercial buildings (e.g., factory, airport). These define areas with high levels of human activity.
Settlement-Agriculture areas include croplands, plantations, and more sparsely settled areas outside urban-dense areas. A sample google earth .kmz file is attached below.
Method: there are various ways to achieve this. Our criterion is that contemporary high resolution imagery such as on google earth be used. Manual digitizing off of google earth imagery can work (that is how we created the sample). The quality of the result is what matters. There is overlap between the classes in the sample; we would use the urban-dense areas to 'cut holes' into the overlapping Settlement-Agriculture areas before merging them.
The project will help inform sustainable development planning in the region.
22 freelancers are bidding on average $1019 for this job
Hello. I'm a GIS expert, Geographer and Cartographer. You can check my previous work in these fields. If you think that I can help you with your project, we can discuss more on chat. Thanks. All the best, Laurentiu.
Hi, I'm an experienced GIS Specialist working in Environmental and Assett Management. I can complete this kind of job using a combination of automatic and manual classification using QGIS.
Hello i am Geodesy student and i am already doing one long term project of digitazing parcels in USA,almost four months. Should i do mabye some example and send you back? I am doing all stuff in ArcGis
I have proven experience in GIS and have done this type of work previously. I will try machine learning algorithms for this task and check output for validation.