def saveCsvBySrc(self, srcBase=None):
if not srcBase:
srcBase = self.rawPath
files = os.listdir(srcBase)
images = []
imgPaths = []
dao = ImageDao()
for index, file in enumerate(files):
fileName = os.path.splitext(file)
src = srcBase + file
type = int(fileName) % 10
image = ImageInfo(str(index).zfill(5), file, src, type)
images.append(image)
imgPaths.append(src)
dao.insertImg(images)
images = []
shutil.rmtree('DataSetRaw')
os.mkdir('DataSetRaw')
for index, newPath in enumerate(imgPaths):
print "add " + str(index) + " picture " + newPath
img = self.loadImage(newPath)
images.append(img)
if len(images) == self.picturePerFile or index == (len(imgPaths) -
1):
dataFileName = 'DataSetRaw/data' + str(
index / self.picturePerFile) + '.csv'
np.savetxt(dataFileName, images, delimiter=',')
images = []
def assessWithoutDist(self):
print "reading data..."
pictures = readCsv(self.dataBaseUrl + "data", self.totalNum)
dao = ImageDao()
imgs = dao.getAll()
typeDict = {}
for img in imgs:
typeDict[img.imgId] = img.imgType
print "training..."
trainSet = pictures[:self.trainNum]
self.knn.train(trainSet)
testSet = pictures[self.trainNum:self.totalNum]
accuracyList = []
heads = ['distance', 'accuracy', 'averageK']
print "predicting..."
for d in range(2000, 4000, 20):
accuracy, avgK = self.knn.predictForManyWithDist(
testSet, self.trainNum, d, typeDict)
item = [d, accuracy, avgK]
accuracyList.append(item)
print "distance:%d accuracy:%f%% averageK:%f" % (
d, accuracy * 100, avgK)
saveCsv(self.resultBasePath + 'assessDist_Radius10_5000-1000.csv',
heads, accuracyList)
def assess(self, k=None):
if not k:
k = self.k
print "reading data..."
pictures = readCsv(self.dataBaseUrl + "data", self.totalNum)
dao = ImageDao()
imgs = dao.getAll()
typeDict = {}
for img in imgs:
typeDict[img.imgId] = img.imgType
print "training..."
trainSet = pictures[:self.trainNum]
self.knn.train(trainSet)
testSet = pictures[self.trainNum:self.totalNum]
zerNp = np.zeros([k, self.testNum])
testLabel = np.arange(self.trainNum, self.totalNum)
for i in range(len(testLabel)):
testLabel[i] = typeDict[str(testLabel[i]).zfill(5)]
testLabel = (zerNp + testLabel).astype('int').T
print "predicting..."
accuracy, avgCriDist = self.knn.predictForManyWithK(
testSet, testLabel, k, typeDict)
print "accuracy:%f%% averageCriticalDist:%f" % (accuracy * 100,
avgCriDist)
def saveCsvWithGaussianBlur(self, srcBase=None, tarBase=None, radius=None):
if not srcBase:
srcBase = self.rawPath
if not tarBase:
tarBase = self.newPath
if not radius:
radius = self.radius
files = os.listdir(srcBase)
images = []
gauImgPaths = []
dao = ImageDao()
shutil.rmtree(tarBase[:-1])
os.mkdir(tarBase[:-1])
for index, file in enumerate(files):
fileName = os.path.splitext(file)[0]
src = srcBase + file
print "blur", index, "picture", src
image = Image.open(src)
image = image.filter(MyGaussianBlur(radius=radius))
tar = tarBase + file
image.save(tar)
type = int(fileName) % 10
image = ImageInfo(str(index).zfill(5), file, src, type)
images.append(image)
gauImgPaths.append(tar)
dao.insertImg(images)
images = []
shutil.rmtree('DataSet')
os.mkdir('DataSet')
for index, newPath in enumerate(gauImgPaths):
print "add", str(index), "picture", newPath
img = self.loadImage(newPath)
images.append(img)
if len(images) == self.picturePerFile or index == (
len(gauImgPaths) - 1):
dataFileName = 'DataSet/data' + str(
index / self.picturePerFile) + '.csv'
np.savetxt(dataFileName, images, delimiter=',')
images = []
def assessWithoutRadius(self, k=None):
if not k:
k = self.k
accuracyList = []
heads = ['radius', 'accuracy', 'averageCriticalDist']
for radius in range(5, 15):
print radius, ':'
print "blurring ..."
data = MyData()
data.saveCsvWithGaussianBlur(radius=radius)
dao = ImageDao()
imgs = dao.getAll()
typeDict = {}
for img in imgs:
typeDict[img.imgId] = img.imgType
zerNp = np.zeros([k, self.testNum])
testLabel = np.arange(self.trainNum, self.totalNum)
for i in range(len(testLabel)):
testLabel[i] = typeDict[str(testLabel[i]).zfill(5)]
testLabel = (zerNp + testLabel).astype('int').T
pictures = readCsv(self.dataBaseUrl + "data", self.totalNum)
print "training..."
trainSet = pictures[:self.trainNum]
self.knn.train(trainSet)
testSet = pictures[self.trainNum:self.totalNum]
print "predicting..."
accuracy, avgCriDist = self.knn.predictForManyWithK(
testSet, testLabel, k, typeDict)
item = [radius, accuracy, avgCriDist]
accuracyList.append(item)
print "k:%d radius:%f accuracy:%f%% averageCriticalDist:%f" % (
k, radius, accuracy * 100, avgCriDist)
saveCsv(self.resultBasePath + 'assessRadiusK' + str(k) + '.csv', heads,
accuracyList)
def assessFeaWithBlurWithoutK(self):
print "reading neaFeaData..."
netFea = readCsv(self.dataBaseUrl + "netFea", self.totalNum) * 256 * 40
print "reading data..."
pictures = readCsv(self.dataBaseUrl + "data", self.totalNum)
pictures = np.append(pictures, netFea, axis=1)
print pictures.shape
dao = ImageDao()
imgs = dao.getAll()
typeDict = {}
for img in imgs:
typeDict[img.imgId] = img.imgType
print "training..."
trainSet = pictures[:self.trainNum]
self.knn.train(trainSet)
testSet = pictures[self.trainNum:self.totalNum]
accuracyList = []
heads = ['k', 'accuracy', 'averageCriticalDist']
print "predicting..."
for k in range(1, 101):
zerNp = np.zeros([k, self.testNum])
testLabel = np.arange(self.trainNum, self.totalNum)
for i in range(len(testLabel)):
testLabel[i] = typeDict[str(testLabel[i]).zfill(5)]
testLabel = (zerNp + testLabel).astype('int').T
accuracy, avgCriDist = self.knn.predictForManyWithK(
testSet, testLabel, k, typeDict)
item = [k, accuracy, avgCriDist]
accuracyList.append(item)
print "k:%d accuracy:%f%% averageCriticalDist:%f" % (
k, accuracy * 100, avgCriDist)
saveCsv(self.resultBasePath + 'assessFeaWithBlurWithoutK.csv', heads,
accuracyList)
def readCsv(baseFileName, dataNum, picturePerFile=1000):
if dataNum == 0:
return None
if dataNum == -1:
dao = ImageDao()
dataNum = dao.getImgCount()
print 'image count', dataNum
fileName = baseFileName + str(0) + ".csv"
print "reading" + fileName
imgs = np.loadtxt(open(fileName, "rb"), delimiter=",", skiprows=0)
fileNum = (dataNum - 1) / picturePerFile + 1
for item in range(1, fileNum):
fileName = baseFileName + str(item) + ".csv"
print "reading" + fileName
data = np.loadtxt(open(fileName, "rb"), delimiter=",", skiprows=0)
imgs = np.append(imgs, data, axis=0)
imgs = imgs[:dataNum]
return imgs
k = 0
optimalDist = 2200
optimalLeastK = 12
dataNum = -1
radius = 10
print "reading data..."
data = MyData()
pictures = readCsv(dataBaseUrl + "data", dataNum)
print "training..."
knn = KNearestNeighbor()
knn.train(pictures)
dao = ImageDao()
imgs = dao.getAll()
typeDict = {}
for img in imgs:
typeDict[img.imgId] = img.imgUrl
# server.config['UPLOAD_FOLDER'] = os.getcwd()
@server.route('/', methods=['GET', 'POST'])
def home():
print "start..."
pictureURL = url_for('static', filename=imgBaseURL + "empty.jpg")
return render_template('main.html',
picture=pictureURL,
select=select,