Use matlab to do a simple project

1. Fine tune the blob detector (implicitly also the edge detector ) to analyze the gray scale version of the image, detect the orange cone or grey trash barrel in the image. 2. Repeat problem one for a HSV component of the image 3. Take many pixel samples taken from the yellow pillar in the image to form a population P1. Repeat for pixels not on the yellow pillar to form a population P2. Design a linear discriminator to classify pixels in the image into either being from population P1 or P2. How well does it work? 4. Fine a binary classifier in question 3 so that it works just as well in at least to of the following images.