Initial commit

Divers/tutoriel37/README.md

@@ -0,0 +1,6 @@
+# Tutoriel 37
+## Local Binary Pattern
+
+La vidéo de ce tutoriel est disponible à l'adresse suivante: https://www.youtube.com/watch?v=zUdEaE8Kpac
+
+

Divers/tutoriel37/camera.py

@@ -0,0 +1,82 @@
+from skimage import feature
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+
+method_distance=[cv2.HISTCMP_CORREL,
+		 		 cv2.HISTCMP_CHISQR,
+				 cv2.HISTCMP_INTERSECT,
+				 cv2.HISTCMP_BHATTACHARYYA,
+				 cv2.HISTCMP_HELLINGER,
+				 cv2.HISTCMP_CHISQR_ALT,
+				 cv2.HISTCMP_KL_DIV]
+
+method_lbp=['default',
+			'ror',
+			'uniform',
+			'var']
+
+cap=cv2.VideoCapture(0)
+numPoints=24
+radius=3
+image_ref=None
+
+width=320
+height=240
+
+windowsize_r=30
+windowsize_c=30
+
+rows=int(height/windowsize_r)
+cols=int(width/windowsize_c)
+
+tab_score=np.empty((rows, cols), dtype=np.float32)
+id_method_distance=6
+id_method_lbp=2
+seuil=0.3
+
+while True:
+	ret, frame=cap.read()
+	image=cv2.resize(frame, (width, height))
+
+	if image_ref is not None:
+		gray=cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+		for r in range(rows):
+			for c in range(cols):
+				window=gray[r*windowsize_r:(r+1)*windowsize_r, c*windowsize_c:(c+1)*windowsize_c]
+				lbp=feature.local_binary_pattern(window, numPoints, radius, method=method_lbp[id_method_lbp])
+				hist, _=np.histogram(lbp, normed=True, bins=numPoints, range=(0, numPoints))
+				score=cv2.compareHist(hist.astype(np.float32), hist_ref.astype(np.float32), method_distance[id_method_distance])
+				tab_score[r, c]=score
+
+	if image_ref is not None:
+		tab_score2=np.zeros_like(tab_score)
+		tab_score2[tab_score<seuil]=1
+		cv2.imshow("SCORE2", cv2.resize(tab_score2, (int(width), int(height)), interpolation = cv2.INTER_NEAREST))
+		tab_score=np.clip(1/(10*tab_score), 0., 1.)
+		cv2.imshow("Score", cv2.resize(tab_score, (int(width), int(height)), interpolation = cv2.INTER_NEAREST))
+
+	cv2.rectangle(frame, (0, 0), (frame.shape[1], 30), (100, 100, 100), cv2.FILLED)
+	txt="[q]Quit   [p|m]Seuil={:2.1f} [d]Methode distance:{}   [l]Methode LBP:{}".format(seuil, method_distance[id_method_distance], method_lbp[id_method_lbp])
+	cv2.putText(frame, txt, (20, 20), cv2.FONT_HERSHEY_PLAIN, 1, (255, 255, 255), 1)
+
+	cv2.imshow("Image", frame)
+
+	key=cv2.waitKey(1)&0xFF
+	if key==ord('q'):
+		quit()
+	if key==ord('i'):
+		image_ref=cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+		lbp_ref=feature.local_binary_pattern(image_ref, numPoints, radius, method=method_lbp[id_method_lbp])
+		hist_ref, _=np.histogram(lbp_ref, normed=True, bins=numPoints, range=(0, numPoints))
+	if key==ord('l'):
+		id_method_lbp=(id_method_lbp+1)%len(method_lbp)
+		if image_ref is not None:
+			lbp_ref=feature.local_binary_pattern(image_ref, numPoints, radius, method=method_lbp[id_method_lbp])
+			hist_ref, _=np.histogram(lbp_ref, normed=True, bins=numPoints, range=(0, numPoints))
+	if key==ord('d'):
+		id_method_distance=(id_method_distance+1)%len(method_distance)
+	if key==ord('p'):
+		seuil+=0.1
+	if key==ord('m'):
+		seuil=max(0.1, seuil-0.1)

Divers/tutoriel37/compare_textures.py

@@ -0,0 +1,64 @@
+from skimage import feature
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+import glob
+
+method_distance=[cv2.HISTCMP_CORREL,
+				 cv2.HISTCMP_CHISQR,
+			  	 cv2.HISTCMP_INTERSECT,
+				 cv2.HISTCMP_BHATTACHARYYA,
+				 cv2.HISTCMP_HELLINGER,
+				 cv2.HISTCMP_CHISQR_ALT,
+				 cv2.HISTCMP_KL_DIV]
+
+method_lbp=['default',
+			'ror',
+			'uniform',
+			'var']
+
+numPoints=8
+radius=3
+id_method_distance=6
+id_method_lbp=2
+
+textures=['gazon', 'gravier', 'bois']
+
+tab_images=[]
+tab_hist=[]
+for texture in textures:
+	print(">>>", "{}.jpg".format(texture))
+	image=cv2.imread("{}.jpg".format(texture), 0)
+	if image is None:
+		quit("Probleme image...")
+	lbp=feature.local_binary_pattern(image, numPoints, radius, method=method_lbp[id_method_lbp])
+	hist_ref, _=np.histogram(lbp, normed=True, density=True, bins=2**numPoints, range=(0, 2**numPoints))
+	tab_images.append(image)
+	tab_hist.append(hist_ref)
+
+for fichier in glob.glob("textures/*.jpg"):
+	print("Lecture de", fichier)
+	frame=cv2.imread(fichier)
+
+	if frame is None:
+		print("Probleme image ...")
+		continue
+
+	cv2.imshow("Image", frame)
+
+	gray=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+	tab_distance=[]
+	for id in range(len(tab_hist)):
+		h=tab_hist[id]
+		lbp=feature.local_binary_pattern(gray, numPoints, radius, method=method_lbp[id_method_lbp])
+		hist, _=np.histogram(lbp, normed=True, density=True, bins=2**numPoints, range=(0, 2**numPoints))
+		score=cv2.compareHist(hist.astype(np.float32), h.astype(np.float32), method_distance[id_method_distance])
+		tab_distance.append(score)
+		print("  score {:10}: {:2.6f}".format(textures[id], score))
+
+	tab_distance=np.array(tab_distance)
+	print("  -> texture:", textures[np.argmin(tab_distance)])
+
+	key=cv2.waitKey()&0xFF
+	if key==ord('q'):
+		quit()

Divers/tutoriel37/lbp.py

@@ -0,0 +1,41 @@
+from skimage import feature
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+
+method_lbp=['default',
+			'ror',
+			'uniform',
+			'var']
+
+cap=cv2.VideoCapture(0)
+numPoints=24
+radius=3
+id_method_lbp=0
+
+while True:
+	ret, frame=cap.read()
+
+	image=cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+	lbp=feature.local_binary_pattern(image, numPoints, radius, method=method_lbp[id_method_lbp])
+
+	cv2.rectangle(frame, (0, 0), (frame.shape[1], 30), (100, 100, 100), cv2.FILLED)
+	txt="[q] Quit  [o|l]numPoints:{:d}  [i|k]rayon:{:d} [m] Methode: {}".format(numPoints, radius, method_lbp[id_method_lbp])
+	cv2.putText(frame, txt, (20, 20), cv2.FONT_HERSHEY_PLAIN, 1, (255, 255, 255), 1)
+
+	cv2.imshow("Image", frame)
+	cv2.imshow("LBP", lbp/np.max(lbp))
+
+	key=cv2.waitKey(1)&0xFF
+	if key==ord('m'):
+		id_method_lbp=(id_method_lbp+1)%len(method_lbp)
+	if key==ord('i'):
+		radius=radius+1
+	if key==ord('k'):
+		radius=max(3, radius-1)
+	if key==ord('o'):
+		numPoints=numPoints+1
+	if key==ord('l'):
+		numPoints=max(3, numPoints-1)
+	if key==ord('q'):
+		quit()

Divers/tutoriel37/lbp_hist.py

@@ -0,0 +1,27 @@
+from skimage import feature
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+import glob
+
+numPoints=8
+radius=3
+
+textures=['gazon', 'gravier', 'bois']
+
+for texture in textures:
+	print(">>>", "{}.jpg".format(texture))
+	image=cv2.imread("{}.jpg".format(texture), 0)
+	if image is None:
+		quit("Probleme image...")
+	lbp=feature.local_binary_pattern(image, numPoints, radius, method='default')
+	hist_ref, _=np.histogram(lbp, bins=2**numPoints, range=(0, 2**numPoints))
+
+	cv2.imshow("Image", image)
+	cv2.imshow("LBP", lbp)
+	plt.plot(hist_ref)
+	plt.show()
+
+	key=cv2.waitKey(1)&0xFF
+	if key==ord('q'):
+		quit()