Thesis Research
Thought I would share this video that was shared by NASAs SDO facebook page, I think it gives alot of background into the importance of solar research as well as it has some great info, visuals and history about solar weather. Enjoy!
In a meeting with my advisers, it was said by one of their colleagues that the AIA 211 images might be more detailed in the features to detect. Since originally I decided to use AIA 193, because visually it was easier for me to see and identify the solar cavities, I decided to try the […]
Continue reading about Running Classifier on Composite Validation Set
Now that a decent Hit Rate has been achieved, the focus turns to reducing the False Alarms. One step to reducing the false alarms is to eliminate overlapping regions of interest, so that the false alarms picked up in about the same vicinity will count as 1 or 2 false alarms as opposed to 6 […]
Continue reading about Trial ::: haarcascade21 ::: [Reduction of overlapping ROIs]
Now that a decent Hit Rate has been achieved, the focus turns to reducing the False Alarms. One step to reducing the false alarms is to black out the suns disk, so that the false alarms picked up in the suns disk will eliminated from the count. Original False Alarm count was 95058, after some […]
Continue reading about Trial ::: haarcascade21 ::: [Black Out the Suns disk]
In testing the classifiers produced, I ran it initially against the training set that was used to create the classifier, as seen in the previous post. So now we are going to start running tests on new image sets, to see how it truly performs on images it has never “seen” before. The below hit […]
Through this process I have been trying to increase my sample sets, so I am currently up to about 3600 positive images resulting from rotating one positive image and tracking the ROI (region of interest) across all subsequent images, as well as increasing the negative set. Below are the commands and parameters used to create […]
When creating samples from the marked images the default options are listed below. You notice that the default for the maxzangle is less than the defaults for the x and y angle values. So why is that? What do these parameters mean? The Haar Classifier was initially developed for facial detections, but has been […]
Continue reading about CreateSamples ::: [ Maxxangle, Maxyangle, Maxzangle ]
In the previous post Advanced Object Marker ::: [Tracking defined ROI (Region Of Interest)] the second video shows the new region of interest (ROI) as the YELLOW rotated rectangle. For our training purposes, a rotated rectangle can not be input to the training application as it expects the defining members of a rectangle as x, […]
Continue reading about Bounding Rectangle for ROI of Rotated Rectangle
Now that we have eliminated scaling from our rotation images, we can now try to track/follow our region of interest (ROI) throughout all rotations. Below is the first initial attempt, we marked our first ROI so we know the left most point of the rectangle as well as its dimensions (width and height). So we […]
Continue reading about Advanced Object Marker ::: [Tracking defined ROI (Region Of Interest)]
So initially we thought to rotate the image 1 degree and save the resulting image, we then ended up with about 360 image files derived from a single image. If we marked just the initial image (non-rotated) then based off of the placement in that image we could derive the position of interest in all […]
Continue reading about Advanced Object Marker ::: [Rotating Images (Rotation and Scaling)]