Info

ldm.py 2018-05-25

Author: Zhao Mingming <471106585@qq.com>

Copyright: This module has been placed in the public domain.

version:0.1.0 - add compare2dir

version:0.1.1 - add face_rec_ms version:0.1.3 - add compare2imglist

Classes: - LDM: you can use a function to get landmarks and face feature with no other libs

Functions:

• face_feature: the feature of face in the image for face recognition

• landmarks: get the landmarks and face in the image

• face_rec: return the face similarity of the face in two images

• has_same_person: return a int(>0) if there have a same person in two images,otherwise 0

• compare2dir: compare the face in the images of the two dirs,and return the rate ,score_avg,class_id1,class_id2

• compare2imglist: compare the face in the imglist ,and return the rate ,score_avg,class_id1,class_id2

How To Use This Module

1. example code:

imagepath="closed_eye/9.jfif"
img=io.imread(imagepath)
img1=img
imagepath="closed_eye/14.jfif"
img2=io.imread(imagepath)

ldmer=ldm.LDM()
print img.shape[0]
ldl,facel,txt=ldmer.landmarks(img)
print txt

for ld in ldl:
    print 10*'-'
    print 'nose:'
    print ld['nose']
for face in facel:
    print 10*'-'
    print 'face:'
    print face.top()
    print face.left()
    print face.width()
    print face.height()
    print face.bottom()
    print face.right()
    x,y,w,h=[face.top(),face.left(),face.width(),face.height()]
    print x,y,w,h
print "feature:"
ffl=ldmer.face_feature(img,facel)
for ff in ffl:
    print help(ff)
    print 'ff='+str(ff)
    print 'len(ff)='+str(len(ff))
    print 'ff[0]='+str(ff[0])
    print 'ff[127]='+str(ff[127])

print "face compare:"
print ldmer.face_rec(img1,img2)
print ldmer.has_same_person(img1,img2)
print ldmer.has_same_person(img2,img2)

print "face number:"
print ldmer.face_number(img,facel)

Refresh

add a function : ldmer.imread(imgpath) modify the return value number to: has_flag,max_score=ldmer.has_same_person(img1,img2)