def getWeightSumTest():
from algorithm import getLbp, getVectors, getWeightSum
labelMap = getSlic(img, 200)
maxLabel = labelMap.max() + 1
im = sk.color.rgb2lab(img)
degreeVectors, Ws = getVectors(im, labelMap)
vectors = getWeightSum(labelMap, degreeVectors, Ws)
m, n = labelMap.shape
imgg = np.zeros((m, n))
imgg2 = np.zeros((m, n))
order = [
'lab', 'l', 'a', 'b', 'lab-texture', 'l-texture', 'a-texture',
'b-texture'
]
labs = [im] + [im[:, :, i] for i in range(3)]
lbps = map(lambda c: getLbp(c, labelMap, 1)[1], labs)
labLbp = labs + lbps
for color in range(vectors.shape[1]):
for k in range(maxLabel):
imgg[labelMap == k] = vectors[k][color]
imgg2[labelMap == k] = degreeVectors[k][color]
print order[color], 'raw | scatter | weight sum'
# show(sk.exposure.equalize_hist(imgg))
show([labLbp[color], imgg2, imgg], 1)
loga(degreeVectors)
loga(vectors)
def grabCutTest():
coarseMethods = ['MY4','QCUT','DRFI']
imgInd = 1
n_segments,compactness = 200,10
imgName = IMG_NAME_LIST[imgInd]
img,imgGt = readImg(imgName)
coarseDic = getCoarseDic(imgName,coarseMethods)
refinedImg = coarseDic['DRFI']
mask = grabCut(img,refinedImg)
imgCut = img*mask[:,:,np.newaxis]
show([img,imgCut])
show([refinedImg,mask])
def getEdgeNeighborTest():
from algorithm import getEdge,getNeighborMatrix,getNeighbor
labelMap = getSlic(img,200)
show(mark_boundaries(img,labelMap),0)
edge = getEdge(labelMap)
neighborMatrix = getNeighborMatrix(labelMap)
m, n = labelMap.shape
dic = getNeighbor(0,labelMap,neighborMatrix,4)
print dic
imgg = np.zeros((m,n))
for k in dic:
imgg[labelMap==k]=dic[k]
show(imgg)
imgg = np.zeros((m,n))
for k in edge:
imgg[labelMap==k]=1
show(imgg)
edge = getEdge(labelMap,0.06)
imgg = np.zeros((m,n))
for k in edge:
imgg[labelMap==k]=1
print 'edge width 0.06'
show(imgg)
def integratImgsBy3wayTest():
from algorithm import getSlic,getCoarseDic,integratImgsBy3way,buildMethodDic
labelMap = getSlic(rgbImg,300)
img = rgbImg
m, n = labelMap.shape
coarseMethods = ['GT']
coarseDic = getCoarseDic(imgName,coarseMethods)
# show(coarseDic)
coarseImgs=coarseDic.values()
g()
refinedImgs=map(lambda f:f(img,coarseImgs,labelMap),buildMethodDic.values())
img = integratImgsBy3way(refinedImgs)
show(img)
def getEdgeImg(img, labelMap=None, width=0.0):
'''
width(float):how width of edge
return a list of label: edge of labelMap
'''
labelMap = getSlic(img, 300) if labelMap is None else labelMap
width = int(min(*labelMap.shape) * width)
u, d, l, r = (labelMap[0:width + 1].ravel(), labelMap[-1 - width:].ravel(),
labelMap[:,
0:width + 1].ravel(), labelMap[:,
-1 - width:].ravel())
edge = np.unique(np.c_[[u], [d], [l], [r]])
new = np.zeros(img.shape)
for i in edge:
new[labelMap == i] = 1
show(new)
return new
def aucBug():
# from saliency import *
# from algorithm import *
# from analysis import *
ll = [('184.jpg', 0.83881506713016329), ('138.jpg', 0.84727243912368833),
('122.jpg', 0.88417034903928537), ('153.jpg', 0.92387361400695434)]
name = ll[0][0]
imgGt = io.imread(IMG_DIR + name[:-3] + 'png').astype(np.bool) #.ravel()
methods = ['DISC2', 'DRFI', 'GMR', 'MEAN']
method = 'DISC2'
coarseDic = getCoarseDic(name, methods, COARSE_DIR)
print coarseDic.keys()
# show(coarseDic.values())
resoult = coarseDic[method]
show([resoult])
show(imgGt)
##imgGt.ravel()
#resoult.ravel()
p, r, _ = precisionRecallCurve(imgGt, resoult)
#l = zip(resoult,imgGt)
print method, 'auc', metrics.roc_auc_score(imgGt.ravel(), resoult.ravel())
plt.plot(p, r)
plt.show()
def my4test():
#if 1:
from algorithm import *
labelMap = getSlic(rgbImg,200)
maxLabel = labelMap.max()+1
m, n = labelMap.shape
coarseMethods = ['MEAN']
coarseDic = getCoarseDic(imgName,coarseMethods)
# show(coarseDic)
sumCoarseImg = getSumCoarseImg(coarseDic)
coarseImgs=coarseDic.values()
img = sk.color.rgb2lab(rgbImg)
# 获得4+4维 distance
degreeVectors, Ws = getVectors(img, labelMap)
weightSumVectors = getWeightSum(labelMap, degreeVectors, Ws)
diffEdges,diffNeighbors = getAllDiffEdgeAndNeighbor(labelMap,Ws)
vectors = np.append(weightSumVectors,diffEdges,1)
vectors = np.append(vectors,diffNeighbors,1)
vectorsTrains = []
coarseTrains = []
for coarseImg in coarseImgs:
coarseTrain, vectorsTrainTag = getCoarseTrain(coarseImg, labelMap)
vectorsTrains += list(vectors[vectorsTrainTag])
coarseTrains += list(coarseTrain)
elm = getElm(np.array(vectorsTrains), np.array(coarseTrains))
refined = elm.predict(vectors)[:,0]
refinedImg = valueToLabelMap(labelMap,normalizing(refined))
show(mark_boundaries(sk.color.lab2rgb(img),labelMap))
print diffEdges.shape
show(valueToLabelMap(labelMap,diffEdges.sum(1)))
show(valueToLabelMap(labelMap,diffNeighbors[:,:4].sum(1)))
# show(valueToLabelMap(labelMap,diffNeighbors[:,4:].sum(1)))
# show(valueToLabelMap(labelMap,diffNeighbors.sum(1)))
show(valueToLabelMap(labelMap,vectors[:,4:8].sum(1)))
def getRefindImgsTest():
IMG_DIR = r'E:\3-experiment\SalBenchmark-master\Data\DataSet1\Imgs/'
COARSE_DIR =r'E:\3-experiment\SalBenchmark-master\Data\DataSet1\Saliency/'
IMG_DIR = 'test/'
COARSE_DIR ='test/'
IMG_NAME_LIST = filter(lambda x:x[-3:]=='jpg',listdir(IMG_DIR))
coarseMethods = ['QCUT','DRFI']
coarseMethods = ['GT']
imgInd = 0
n_segments,compactness = 200,10
imgName = IMG_NAME_LIST[imgInd]
img,imgGt = readImg(imgName)
coarseDic = getCoarseDic(imgName,coarseMethods)
#show(coarsesDic)
sumCoarseImg = getSumCoarseImg(coarseDic)
coarseImgs=coarseDic.values()
labelMap = getSlic(img,n_segments,compactness)
rgb = img
img = sk.color.rgb2lab(img)
#show([mark_boundaries(img,labelMap),imgGt])
# 获得4+4维 distance
degreeVectors, Ws = getVectors(img, labelMap)
vectors = getWeightSum(labelMap, degreeVectors, Ws)
vectorsTrains = []
coarseTrains = []
for coarseImg in coarseImgs:
coarseTrain, vectorsTrainTag = getCoarseTrain(coarseImg, labelMap)
vectorsTrains += list(vectors[vectorsTrainTag])
coarseTrains += list(coarseTrain)
elm = getElm(np.array(vectorsTrains), np.array(coarseTrains))
refined = elm.predict(vectors)[:,0]
refinedImg = valueToLabelMap(labelMap,normalizing(refined))
vectorsImg = valueToLabelMap(labelMap,normalizing(vectors.sum(1)))
show([rgb,refinedImg])
show([rgb,vectorsImg])
show(vectorsImg-refinedImg)
loga(vectorsImg-refinedImg)
g()
COARSE_DIR = '../DataSet1/Saliency/'
IMG_DIR = r'G:\Data\HKU-IS/Imgs/'
COARSE_DIR = r'G:\Data\HKU-IS/Saliency/'
IMG_NAME_LIST = filter(lambda x: x[-3:] == 'jpg', listdir(IMG_DIR))
setModuleConstant(alg)
imgName = IMG_NAME_LIST[3]
img, imgGt = readImg(imgName)
rgbImg = img
rgbImg = np.zeros((100, 100, 3))
rgbImg[25:75, 25:75, 1:] = 1.
show(rgbImg)
def integratImgsBy3wayTest():
from algorithm import getSlic, getCoarseDic, integratImgsBy3way, buildMethodDic
labelMap = getSlic(rgbImg, 300)
img = rgbImg
m, n = labelMap.shape
coarseMethods = ['MEAN']
coarseDic = getCoarseDic(imgName, coarseMethods)
# show(coarseDic)
coarseImgs = coarseDic.values()
refinedImgs = map(lambda f: f(img, coarseImgs, labelMap),
buildMethodDic.values())