Spatial visualization is the ability of an individual to visualize an object mentally and understand its spatial orientation. It plays an important role in engineering. There have been multiple works that show that spatial visualization skills can be improved with the right training. Creating technical drawings requires mental manipulation and visual thinking. Strong spatial visualization abilities are required for such a mental process. To train students in this process and to improve their spatial skills, Professor Diana Bairaktarova conducted a class in spatial visualization for freshmen in engineering at Virginia Polytechnic Institute and State University. The class consisted of 169 students from different engineering disciplines. One of the major pedagogical techniques used in the class was to give students three assignments on drawing sectional views of mechanical objects with a pre-defined cutting plane. All three assignments had the same six mechanical objects to assess their improvement through the class. Students were not given feedback after completion of each assignment. They were asked to do the first assignment at the beginning of the course, the second assignment during the middle of the semester and the final assignment towards the end of the semester. The students were given instructions on how to draw the 2D sectional views. The assignments were then graded by the teaching assistants for the class and the improvement of the students through the semester was recorded. Assignments were graded manually by two different raters using a pre-defined grading rubric. An inter-rater reliability was established between the graders. There were drawbacks to this process. It was extremely time consuming since there were more than 500 assignments to be graded by the teaching assistants. Also, to establish inter-rater reliability, the assignments had to be graded twice. The process would have been more efficient if there was a software that could automate the grading process. Also, this would eliminate the need to establish an inter-rater reliability. This research aims at developing a software for automatic grading of the technical drawings. The software gives students' feedback on the drawings describing their mistakes. This would give a more complete learning experience as the students would get a better understanding of the internal details of the object with the help of the feedback they are getting. In addition, the software is independent of the CAD platform used to create the drawings. The instructor can also upload a batch of images that can be processed by the software at once. The grading rubric that was used for manual grading can be implemented in the software. The software uses Image processing and Computer Vision toolboxes in MATLAB which enables the comparison between the submitted technical drawing and the source (solution) drawing. The software is currently developed for simple geometries with less complicated features as it is being employed in a course where students are new to CAD environment. Students can also use this software as an interactive learning tool as they create 3D models and sectional views of mechanical objects. The proposed software reduces the amount of effort put in by faculty on grading the assignments. It also gives students feedback on the drawings, making it an interactive tool which improves the learning experience. This software can be a powerful pedagogical tool to improve spatial visualization skills. / Master of Science