As a result of the massive development of computing and information technology,
computer vision systems are becoming more and more common in the industrial area,
including electrical manufacturing. Nowadays, AOI (Automated Optical Inspection)
machines are essential parts of the manufacturing technology.
During my thesis I was working on the Mirtec MV-9 AOI machine. The main
goal of my thesis was to analyze and test the image processing algorithms of this platform
that are not yet used during the manufacturing phase and to implement them in the AOI
In the introduction part of my thesis I will demonstrate the electrical
manufacturing processes and introduce the reader to the AOI machines in general.
In the following part I will present the Mirtec MV-9 AOI platform’s structure,
way of operation and the image processing software environment.
In the remaining part of my thesis I will showcase the activities and procedures
that I performed on each image processing algorithm in a way described below.
As the first step for each image processing algorithm, I performed an analysis and
study, during which I got familiar with the functioning and operating of the algorithm.
The next step was to find a possible solution in which the given algorithm could be used
for manufacturing error detection. This was followed by the testing procedure, where I
analyzed several test cases in order to determine whether the given algorithm could be
reliably used for the previously defined error detection or not. Granted that the test was
successful, the last step was to implement the algorithm in the AOI inspection procedure.
After presenting the procedure that were done for each algorithm, I will
summarize the final results of the thesis.