def kMean(clusterNum):
g = Graph(r"E:\ds2018")
vocaDir = r"temp/vocabulary/"
if not os.path.exists(vocaDir):
os.makedirs(vocaDir)
sift = SIFT()
centers = []
for i in range(g.getTypeNum()):
print("[kmeans]:" + str(i))
imgPaths = g.getTrainSet(i)
features = np.float32([]).reshape((0, 128))
for imgPath, type in imgPaths:
imgMat = g.getGreyGraph(imgPath)
if imgMat is None:
print("[kmean]:" + imgPath + " is None")
continue
feature = sift.getFeature(imgMat)
features = np.append(features, feature, axis=0)
kmeans = KMeans(n_clusters=clusterNum).fit(features)
filename = os.path.join(vocaDir, str(i) + ".npy")
np.save(filename, kmeans.cluster_centers_)
centers.append(kmeans.cluster_centers_)
return centers
if mat is None:
print(imgPath+"is None")
continue
feature=sift.getFeature(mat)
featVec=calcFeatVec(feature,centers[type],clusterNum)
trainData=np.append(trainData,featVec,axis=0)
trainTypes=np.append(trainTypes,np.int32([type]))
print(trainTypes.shape)
trainTypes=trainTypes.astype(np.float32)
'''
print("calc kmeans")
lbp = LBP([8], [1])
hog = HOG()
imgList = g.getTrainSet()
trainTypes, trainDataSift, centers = sift.calcVectorForSvm(
imgList, clusterNum, 1)
print("calc kmeans done")
print("calc lbp")
trainDataLbp = lbp.getFeatVecForSvm(imgList, 0)
print("calc lbp done")
print("calc hog")
trainDataHog = hog.getFeatVecForSvm(imgList, 1)
print("calc hog done")
trainData = np.append(trainDataSift, trainDataLbp, axis=1)
trainData = np.append(trainData, trainDataHog, axis=1)
#trainData=trainDataLbp
self.distance = distance
self.angle = angle
'''
return :list of features
'''
def getFeature(self, imgMat):
re = feature.greycomatrix(imgMat, self.distance, self.angle)
features = []
for i in range(len(self.distance)):
for j in range(len(self.angle)):
features.append(re[:, :, i, j])
return features
g = Graph(r"E:\ds2018")
trainList = g.getTrainSet()
glcm = GLCM([1], [0, np.pi / 4, np.pi / 2, np.pi * 3 / 4])
features = []
print(trainList)
for img in trainList:
#print(img[0])
mat = g.getGreyGraph(img[0])
if mat is None:
print("None")
continue
features.extend(glcm.getFeature(mat))
print(features)
