What did you personally accomplish this week on the project? Give files or photos that demonstrate your progress. Prove to the reader that you put sufficient effort into the project over the course of the week (12+ hours):
Currently ,I am primarily relying on the stats values (left, top coordinates of individual braille dots,as well as the width and height of the neighboring dots) from “cv2.connectedComponentsWithStats()” function. I have checked the exact pixel locations of the spitted out matrices and the original image and have confirmed that the values are in fact accurate. My current redundancies of dots come from the inevitable flaw of the connectedComponentsWithStats() function, and I need to get rid of the redundant dots sporadically distributed in nearby locations using the non_max_suppression. There is a little issue going on, and I do not want to write the whole function myself so I am looking for ways to fix this, but as long as this gets done, I am nearly done with the pre-processing procedures. Currently, I am looking more into how nms filter gets rid of redundant images and am planning on writing the nms filter function myself if it could not be debugged in time. I wasn’t able to make progress as of now (saturday) as far as the non_maximum_suppression filter is concerned, so further works are required tmrw in order to discuss some progress.
Is your progress on schedule or behind? If you are behind, what actions will be taken to catch up to the project schedule?:
The current, pre-defined non_max_suppression is currently not fully functional and am currently in the process of debugging to properly get rid of redundant dots. However, once this is taken care of, the pre-processing subsystem is nearly done so I am hoping to get it solved and done by thanksgiving-ish.
What deliverables do you hope to complete in the next week?:
The current non_max_suppression filter would either be tweaked or re-written to properly process redundant dots sporadically present near the target location. Furthermore, I have currently confirmed that the center coordinates of each braille dots as well as their corresponding width and height are accurate given the direct pixel-by-pixel comparison with the original image, so cropping should be completed with relative ease as well.