required=True,
help="Path to where we stored our index")
ap.add_argument("-q", "--query", required=True, help="Path to query image")
args = vars(ap.parse_args())
# load the query image and show it
queryImage = cv2.imread(args["query"])
cv2.imshow("Query", queryImage)
print("queryImage " + str(queryImage))
print "query: %s" % (args["query"])
# describe the query in the same way that we did in
# index.py -- a 3D RGB histogram with 8 bins per
# channel
desc = RGBHistogram([8, 8, 8])
queryFeatures = desc.describe(queryImage)
# load the index perform the search
index = cPickle.loads(open(args["index"]).read())
searcher = Searcher(index)
results = searcher.search(queryFeatures)
# initialize the two montages to display our results --
# we have a total of 25 images in the index, but let's only
# display the top 10 results; 5 images per montage, with
# images that are 400x166 pixels
montageA = np.zeros(
(len(queryImage) * 5, len(queryImage[0]), 3),
dtype="uint8") #np.zeros((166 * 5, 400, 3), dtype = "uint8")
montageB = np.zeros(
(len(queryImage) * 5, len(queryImage[0]), 3),
help = "Path to where the computed index will be stored")
args = vars(ap.parse_args())
# initialize the index dictionary to store our our quantifed
# images, with the 'key' of the dictionary being the image
# filename and the 'value' our computed features
index = {}
# initialize our image descriptor -- a 3D RGB histogram with
# 8 bins per channel
desc = RGBHistogram([8, 8, 8])
# use glob to grab the image paths and loop over them
for imagePath in glob.glob(args["dataset"] + "/*.jpg"):
# extract our unique image ID (i.e. the filename)
k = imagePath[imagePath.rfind("/") + 1:];print("k "+k)
# load the image, describe it using our RGB histogram
# descriptor, and update the index
image = cv2.imread(imagePath)
features = desc.describe(image)
index[k] = features
# we are now done indexing our image -- now we can write our
# index to disk
f = open(args["index"], "w")
f.write(cPickle.dumps(index))
f.close()
# show how many images we indexed
print "done...indexed %d images" % (len(index))
# use glob to grab the image paths and loop over them
for imagePath in glob.glob(args["dataset"] + os.sep + "*.png"):
# load the image, describe it using our RGB histogram
# descriptor, and update the index
image = cv2.imread(imagePath)
# print(os.path.basename(k))
# Print each image to get the working status
print(imagePath)
print("-------------")
# extract our unique image ID (i.e. the filename)
k = imagePath[imagePath.rfind(os.sep) + 1:]
# print(k)
features = desc.describe(image)
index[k] = features
elif option == "lbp":
print("LBP Descriptor")
# initialize the local binary patterns descriptor along with
# the data and label lists
desc = LocalBinaryPatterns(24, 8)
# data = []
# labels = []
# size = len(glob.glob(args["dataset"] + os.sep + "*.png"))
# Training
# loop over the training images
for imagePath in glob.glob(args["dataset"] + os.sep + "*.png"):
# load the image, convert it to gray scale, and describe it
# initialize the index dictionary to store our our quantifed
# images, with the 'key' of the dictionary being the image
# filename and the 'value' our computed features
index = {}
# initialize our image descriptor -- a 3D RGB histogram with
# 8 bins per channel
desc = RGBHistogram([8, 8, 8])
# use glob to grab the image paths and loop over them
for imagePath in glob.glob(args["dataset"] + "/*.png"):
# extract our unique image ID (i.e. the filename)
k = imagePath[imagePath.rfind("/") + 1:]
# load the image, describe it using our RGB histogram
# descriptor, and update the index
image = cv2.imread(imagePath)
features = desc.describe(image)
index[k] = features
# we are now done indexing our image -- now we can write our
# index to disk
f = open(args["index"], "w")
f.write(cPickle.dumps(index))
f.close()
# show how many images we indexed
print "done...indexed %d images" % (len(index))
# loop over the image and mask paths
#for (imagePath, maskPath) in zip(imagePaths, maskPaths):
# # load the image and mask
# image = cv2.imread(imagePath)
# mask = cv2.imread(maskPath)
# mask = cv2.cvtColor(mask, cv2.COLOR_BGR2GRAY)
for imagePath in imagePaths:
# load the image and mask
image = cv2.imread(imagePath)
# mask = cv2.imread(maskPath)
mask = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# describe the image
features = desc.describe(image, mask)
# update the list of data and targets
data.append(features)
target.append(imagePath.split("_")[-2])
# grab the unique target names and encode the labels
targetNames = np.unique(target)
le = LabelEncoder()
target = le.fit_transform(target)
# construct the training and testing splits
(trainData, testData, trainTarget, testTarget) = train_test_split(data, target,
test_size = 0.3, random_state = 42)
# train the classifier
image = cv2.imread(imagePath)
gray= cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
kp, dsc= sift.detectAndCompute(gray, None)
BOW.add(dsc)
dictionary = BOW.cluster()
desc2 = SIFTHistogram(dictionarySize,dictionary)
# use glob to grab the image paths and loop over them
for imagePath in glob.glob(args["dataset"] + "/*.jpg"):
# extract our unique image ID (i.e. the filename)
k = imagePath[imagePath.rfind("/") + 1:]
# load the image, describe it using our RGB histogram
# descriptor, and update the index
image = cv2.imread(imagePath)
features = list(desc.describe(image))
features2 = desc2.describe(image)
features = features + features2
# features = features2
index[k] = np.array(features)
# we are now done indexing our image -- now we can write our
# index to disk
f = open(args["index"], "w")
f.write(cPickle.dumps(index))
f.close()
# show how many images we indexed
print "done...indexed %d images" % (len(index))
ap.add_argument("-i", "--index", required = True,
help = "Path to where we stored our index")
ap.add_argument("-q", "--query", required = True,
help = "Path to query image")
args = vars(ap.parse_args())
# load the query image and show it
queryImage = cv2.imread(args["query"])
cv2.imshow("Query", queryImage)
print "query: %s" % (args["query"])
# describe the query in the same way that we did in
# index.py -- a 3D RGB histogram with 8 bins per
# channel
desc = RGBHistogram([8, 8, 8])
queryFeatures = desc.describe(queryImage)
# load the index perform the search
index = cPickle.loads(open(args["index"]).read())
searcher = Searcher(index)
results = searcher.search(queryFeatures)
# initialize the two montages to display our results --
# we have a total of 25 images in the index, but let's only
# display the top 10 results; 5 images per montage, with
# images that are 400x166 pixels
montageA = np.zeros((166 * 5, 400, 3), dtype = "uint8")
montageB = np.zeros((166 * 5, 400, 3), dtype = "uint8")
# loop over the top ten results
for j in xrange(0, 10):
directory = args["query"]
for filename in os.listdir(directory):
if filename.endswith(".png"):
queryPath = os.path.join(directory, filename)
print(queryPath)
queryImage = cv2.imread(queryPath)
queryImage = cv2.resize(queryImage, (450, 360))
cv2.putText(queryImage, queryPath, (10, 30), cv2.FONT_HERSHEY_SIMPLEX,
1.0, (0, 0, 255), 3)
cv2.imshow("Query", queryImage)
print("query: %s" % queryPath)
if search == 0:
if args["descriptor"] == "rgb":
desc = RGBHistogram([8, 8, 8])
queryFeatures = desc.describe(queryImage)
elif args["descriptor"] == "lbp":
desc = LocalBinaryPatterns(24, 8)
gray = cv2.cvtColor(queryImage, cv2.COLOR_BGR2GRAY)
queryFeatures = desc.describe(gray)
elif args["descriptor"] == "hog":
winSize = (64, 64)
blockSize = (16, 16)
blockStride = (8, 8)
cellSize = (8, 8)
nbins = 9
derivAperture = 1
winSigma = 4.
histogramNormType = 0
L2HysThreshold = 2.0000000000000001e-01
gammaCorrection = 0
'--dataset',
required=True,
help='Path to the directory that contains the images we just indexed')
ap.add_argument('-i',
'--index',
required=True,
help='Path to where we stored our index')
ap.add_argument('-q', '--query', required=True, help='Path to query imag')
args = vars(ap.parse_args())
path = args['query']
image = cv2.imread(path)
cv2.imshow("Query", image)
desc = RGBHistogram([8, 8, 8])
queryFeature = desc.describe(image)
index = pickle.loads(open(args['index'], 'rb').read())
searcher = Searcher(index)
results = searcher.search(queryFeature)
montageA = np.zeros((166 * 5, 400, 3), dtype='uint8')
montageB = np.zeros((166 * 5, 400, 3), dtype='uint8')
for j in range(0, 10):
(score, imageName) = results[j]
path = os.path.join(args['dataset'], imageName)
result = cv2.imread(path)
print('\t {}. {} : {:.3f}'.format(j + 1, imageName, score))
if j < 5:
montageA[j * 166:(j + 1) * 166, :] = result
else:
# initialize the list of data and class label targets
data = []
target = []
# initialize the image descriptor
desc = RGBHistogram([8, 8, 8])
# loop over the image and mask paths
for (imagePath, maskPath) in zip(imagePaths, maskPaths):
# load the image and mask
image = cv2.imread(imagePath)
mask = cv2.imread(maskPath)
mask = cv2.cvtColor(mask, cv2.COLOR_BGR2GRAY)
# describe the image
features = desc.describe(image, mask)
# update the list of data and targets
data.append(features)
target.append(imagePath.split("_")[-2])
# grab the unique target names and encode the labels
targetNames = np.unique(target)
le = LabelEncoder()
target = le.fit_transform(target)
# construct the training and testing splits
(trainData, testData, trainTarget, testTarget) = train_test_split(data, target,
test_size = 0.3, random_state = 42)
# train the classifier
imagePaths = sorted(glob.glob(args['images'] + "/*.jpg"))
maskPaths = sorted(glob.glob(args["masks"] + "/*.png"))
data = []
target = []
desc = RGBHistogram([8, 8, 8])
for (imagePath, maskPath) in zip(imagePaths, maskPaths):
image = cv2.imread(imagePath)
mask = cv2.imread(maskPath)
mask = cv2.cvtColor(mask, cv2.COLOR_BGR2GRAY)
import pdb
pdb.set_trace()
features = desc.describe(image, mask)
data.append(features)
target.append(imagePath.split("_")[-2])
targetNames = np.unique(target)
le = LabelEncoder()
target = le.fit_transform(target)
(trainData, testData, trainTarget,
testTarget) = train_test_split(data, target, test_size=0.3, random_state=42)
model = RandomForestClassifier(n_estimators=25, random_state=84)
model.fit(trainData, trainTarget)
print(
classification_report(testTarget,
model.predict(testData),
target_names=targetNames))
