Automatiserat system för insamling och distribuering av information skriven på krittavlor
Examensarbete för kandidatexamen
Datateknik 300 hp (civilingenjör)
This paper describes the development of an automatic system for capturing images of blackboards, image processing and publishing the resulting images to a web server. At the time of this paper much of the education was still done with the use of blackboards. During lectures it can be hard to follow the lecturers reasonings and conclusions while trying to write down what has been written on the board. The purpose of this bachelor’s thesis is to make it easier for students by creating an automatic note taking system. The thesis describes the early stages of the project with planning and feasibility studies in order to be able to decide how to best develop the product. The development process describes all the design choices and the different modules like the camera unit and the server. With a working prototype, testing was made on all the different modules of the project to be able to evaluate and assess the safety of the product. The result is later processed and a discussion is made on how the product can be made better in the future. The product resulted in a prototype where all expected basic functionality were fully operational. The product were shown to a number of people and most of them had a positive attitude towards the product. It was decided to use one single-board computer in every lecture hall with one camera for each blackboard column and one sensor for eack blackboard. Raspberry Pi was chosen to be used as the single-board computer. Raspberry Pi Camera Modules and photomicrosensors were chosen as components to implement our prototype. Linux was used as operating system for both the camera unit as well as the server. A blackboard is photographed when it is in top position in its respective column and the resulting image is sent to the server. A web server was created where one can search and find lecture notes. An administrator interface was created to be able to manage camera units, courses and users. Image processing was implemented from scratch through the use of relevant litterature. Image processing incorporates cropping, skew correction, determining if something is written on the blackboard and the ability to compare earlier blackboard images with newly taken ones to reduce the number of duplicates.
Data- och informationsvetenskap , Computer and Information Science