Machine_learning_project

Files in this folder are temporarily for Hallux valgus project(CS542S)

Project7_image.py is the code for imaging processing. I have fixed the error and add desciptions. Please put it next to the h5 database

data_processing3.py is the code to generate h5 format database. After running the code, a file named as Dataset30HV.hdf5 (823.2MB) will be generate.

Xt= np.array(f['/entry/Xt']) #30800800

30 original images Xlt=np.array( f['/entry/Xlt']) #120800800

120 groundtruth images (1 original images produces 4 bone-label images(currently we only study four bone)) Xl=np.array(f['/entry/Xl']) #120,

labels(1,2,3,4,1,2,3,4,...,and so on)

H= np.array(f['/entry/H']) #30100100

100*100 patches of hallux(toe) binary image

PP= np.array(f['/entry/PP']) #30 200 150

200*150 patches of proximal phalanx binary image

M1= np.array(f['/entry/M1']) #(30, 400, 250)

400*250 patches of first metatarsal bone binary image

M2= np.array(f['/entry/M2']) #(30, 450, 200)

450*200 patches of second metatarsal bone binary image