Study on the accuracy of school location information in South Africa

Autor: Anita van Deventer, Sives Govender, Lauren Hankel, Melissa Burgess, Ronel Smith, Merryl Ford, Kobus Roux
Rok vydání: 2017
Předmět:
Zdroj: South African Journal of Geomatics; Vol 6, No 2 (2017): Special Edition; 142-154
ISSN: 2225-8531
DOI: 10.4314/sajg.v6i2.1
Popis: Accurate location information is required for proper planning and informed decision making in a variety of sectors. In the basic education sector, accurate school location information is typically required for road, electricity, internet connectivity and water infrastructure planning as well as planning for the delivery of textbooks and public transport (i.e. busses, taxis). The National Education Collaboration Trust (NECT) commissioned the Council for Scientific and Industrial Research (CSIR) to conduct a study of existing school location information in five NECT education districts (Bohlabela, Bojanala, Mount Frere, Uthungulu and Waterberg). School location information in two existing databases, i.e. Education Management Information System (EMIS) and National Education Information Management System (NEIMS) were assessed. Due to the nature of school buildings (i.e. varying form of schools) it is challenging to automate the identification of schools from satellite imagery by using machine learning/image processing techniques. Manual Geographic Information System (GIS) techniques were applied to conduct the study. High resolution satellite imagery and Google StreetView were utilised to ascertain the locations of schools. This study indicated that there are discrepancies between the EMIS and NEIMS databases and that there is a significant amount of school location information that might not be useful for proper planning and informed decision making in certain sectors due to the degree of positional inaccuracy of the data. If the positional accuracy of the incorrect school location information improves, it will have a positive impact on the overall outcomes of planning and decision making.
Databáze: OpenAIRE