Design and Implementation of Automatic Student Evaluation form Processing System Using Image Analysis Techniques
No Thumbnail Available
Date
2011-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Addis Ababa University (AAU) is one of the largest universities in Ethiopia. It has more than
sixty thousand students and more than two thousand academic staffs. To know the teaching
effectiveness of the academic staff and to get the student views regarding the teaching
performance of instructors, the university has prepared a questionnaire form, which contains
about 36 statements that will be filled by students. These forms are processed manually in such a
way that the values given by the students for each statement are added and the average values are
computed. Then these values are divided by the total number of students to get the total average
values given by all the students.
Processing the student evaluation form manually for a large number of students makes the
processing of the forms tedious, time taking and error prone. To alleviate this problem a
mechanism has to be developed by using different form processing techniques. In this project a
digital image analysis technique is used to process the form.
For this project, MATLAB is used as a programming tool since it is high-performance language
for technical computing that integrates computation, visualization, and programming where
problems and solutions are expressed in mathematical notation. The system is designed to
recognize only the values encircled by the students against each statement. After the document is
scanned a template image is taken from the normalized gradient field of the image. Gradient is
the change in gray level with direction. This can be calculated by taking the difference in value
of neighboring pixels, producing a vector for each pixel. Thus, the gradient field of the image is
normalized with a certain threshold value. Noise is removed while computing the gradient field
of the image. Template matching image analysis technique is used, to detect regions which
contain choices against each statement. Then, values encircled by the students are detected by
finding relative circles having a range of radius in the detected region. The coordinate values of
the detected circles are used to recognize values encircled by the students.
Finally, recognized values for each statement are added and the average values are computed.
Then these values are divided by the total number of students to get the total average values
given by all the students.
Keywords: Addis Ababa University, Student evaluation form, Template matching, Recognition
Description
Keywords
Addis Ababa University; Student Evaluation form Template Matching; Recognition;