我们从Python开源项目中，提取了以下18个代码示例，用于说明如何使用cv2.getTickCount()。
def lipSegment(self, img): # self.t1 = cv2.getTickCount() lipHull = self.dlib_obj.get_landmarks(img) cv2.drawContours(img, lipHull, -1, (255, 0, 0), 2) (x, y), (MA, ma), angle = cv2.fitEllipse(lipHull) a = ma/2 b = MA/2 eccentricity = sqrt(pow(a, 2)-pow(b, 2)) eccentricity = round(eccentricity/a, 2) cv2.putText(img, 'E = '+str(round(eccentricity, 3)), (10, 350), self.font, 1, (255, 0, 0), 1) if(eccentricity < 0.9): self.flags.cmd = 'b' else: self.flags.cmd = 'f' if angle < 80: self.flags.cmd = 'l' elif angle > 100: self.flags.cmd = 'r' cv2.putText(img, 'Cmd = ' + self.flags.cmd, (10, 300), self.font, 1, (0, 0, 255), 1, 16) # self.t2 = cv2.getTickCount() # print "Time = ", (self.t2-self.t1)/cv2.getTickFrequency() return img
def mk_list(jmax,imax,tmx,inc,dem,sang): f_list=[] t_setx=t_set[:,0,0] #r_setx=t_set[:,0,0] r_setx=np.array([0.0,0.2,0.4,0.6,0.8,1.0]) nlen=len(t_setx) s_setx=smin*np.ones(nlen) t=cv2.getTickCount() for j in range(jmax): temp=[] for i in range(imax): frefx=fref(tmx[j,i],inc[j,i],dem[j,i]/1000.0,t_setx,r_setx,s_setx,sang[j,i]) temp=temp+[frefx] if j % 100 == 0: print j,(cv2.getTickCount()-t)/cv2.getTickFrequency() f_list=f_list+[temp] return f_list
def mclass(cls1,cls2,cls3,nmax): num=imax*jmax cls1x=np.float32(cls1.reshape(num)) cls2x=np.float32(cls2.reshape(num)) cls3x=np.float32(cls3.reshape(num)) data=np.array([[cls1x],[cls2x],[cls3x]]) data=data.reshape(3,num) data=np.transpose(data) datax=data[::100,:] t=cv2.getTickCount() codebook, destortion = scipy.cluster.vq.kmeans(datax, nmax, iter=10, thresh=1e-05) print (cv2.getTickCount()-t)/cv2.getTickFrequency() t=cv2.getTickCount() code, dist = scipy.cluster.vq.vq(data, codebook) print (cv2.getTickCount()-t)/cv2.getTickFrequency() return code.reshape(jmax,imax)
def detectTemplateMatching(self, img): self.templateMatchingCurrentTime = cv2.getTickCount() duration = (self.templateMatchingCurrentTime - self.templateMatchingStartTime)/cv2.getTickFrequency() if duration > settings.templateMatchingDuration or self.trackedFaceTemplate[2] == 0 or self.trackedFaceTemplate[3] == 0: self.foundFace = False self.isTemplateMatchingRunning = False return faceTemplate = self.getSubRect(img, self.trackedFaceTemplate) roi = self.getSubRect(img, self.trackedFaceROI) match = cv2.matchTemplate(roi, faceTemplate, cv2.TM_SQDIFF_NORMED) cv2.normalize(match, match, 0, 1, cv2.NORM_MINMAX, -1) minVal, maxVal, minLoc, maxLoc = cv2.minMaxLoc(match) foundTemplate = ( minLoc[0] + self.trackedFaceROI[0], minLoc[1] + self.trackedFaceROI[1], self.trackedFaceTemplate[2], self.trackedFaceTemplate[3]) self.trackedFaceTemplate = foundTemplate self.trackedFace = self.scaleRect(self.trackedFaceTemplate, img, 2) self.trackedFaceROI = self.scaleRect(self.trackedFace, img, 2)
def clock(): return cv2.getTickCount() / cv2.getTickFrequency()
def predict_image(flag): t_start = cv2.getTickCount() config = tf.ConfigProto() # config.gpu_options.per_process_gpu_memory_fraction = 0.9 config.gpu_options.allow_growth = True set_session(tf.Session(config=config)) with open(os.path.join(flag.ckpt_dir, flag.ckpt_name, 'model.json'), 'r') as json_file: loaded_model_json = json_file.read() model = model_from_json(loaded_model_json) weight_list = sorted(glob(os.path.join(flag.ckpt_dir, flag.ckpt_name, "weight*"))) model.load_weights(weight_list[-1]) print "[*] model load : %s"%weight_list[-1] t_total = (cv2.getTickCount() - t_start) / cv2.getTickFrequency() * 1000 print "[*] model loading Time: %.3f ms"%t_total imgInput = cv2.imread(flag.test_image_path, 0) input_data = imgInput.reshape((1,256,256,1)) t_start = cv2.getTickCount() result = model.predict(input_data, 1) t_total = (cv2.getTickCount() - t_start) / cv2.getTickFrequency() * 1000 print "Predict Time: %.3f ms"%t_total imgMask = (result[0]*255).astype(np.uint8) imgShow = cv2.cvtColor(imgInput, cv2.COLOR_GRAY2BGR) _, imgMask = cv2.threshold(imgMask, int(255*flag.confidence_value), 255, cv2.THRESH_BINARY) imgMaskColor = cv2.applyColorMap(imgMask, cv2.COLORMAP_JET) # imgZero = np.zeros((256,256), np.uint8) # imgMaskColor = cv2.merge((imgZero, imgMask, imgMask)) imgShow = cv2.addWeighted(imgShow, 0.9, imgMaskColor, 0.3, 0.0) output_path = os.path.join(flag.output_dir, os.path.basename(flag.test_image_path)) cv2.imwrite(output_path, imgShow) print "SAVE:[%s]"%output_path
def train_visualization_seg(self, model, epoch): image_name_list = sorted(glob(os.path.join(self.flag.data_path,'train/IMAGE/*/*.png'))) print image_name_list image_name = image_name_list[-1] image_size = self.flag.image_size imgInput = cv2.imread(image_name, self.flag.color_mode) output_path = self.flag.output_dir input_data = imgInput.reshape((1,image_size,image_size,self.flag.color_mode*2+1)) t_start = cv2.getTickCount() result = model.predict(input_data, 1) t_total = (cv2.getTickCount() - t_start) / cv2.getTickFrequency() * 1000 print "[*] Predict Time: %.3f ms"%t_total imgMask = (result[0]*255).astype(np.uint8) imgShow = cv2.cvtColor(imgInput, cv2.COLOR_GRAY2BGR) imgMaskColor = cv2.applyColorMap(imgMask, cv2.COLORMAP_JET) imgShow = cv2.addWeighted(imgShow, 0.9, imgMaskColor, 0.4, 0.0) output_path = os.path.join(self.flag.output_dir, '%04d_'%epoch+os.path.basename(image_name)) cv2.imwrite(output_path, imgShow) # print "SAVE:[%s]"%output_path # cv2.imwrite(os.path.join(output_path, 'img%04d.png'%epoch), imgShow) # cv2.namedWindow("show", 0) # cv2.resizeWindow("show", 800, 800) # cv2.imshow("show", imgShow) # cv2.waitKey(1)
def retrieve_data_set(): """Retrieve data from all the .npz files and aggregate it into a data set for mlp training""" start_time = cv2.getTickCount() print("Loading data set...") image_array = np.zeros((1, 38400), 'float') label_array = np.zeros((1, 4), 'float') # Retrieve a list of pathname that matches the below expr data_set = glob.glob("data_set/*.npz") if not data_set: print("No data set in directory, exiting!") sys.exit() for single_npz in data_set: with np.load(single_npz) as data: temp_images = data["images"] temp_labels = data["labels"] image_array = np.vstack((image_array, temp_images)) label_array = np.vstack((label_array, temp_labels)) X = np.float32(image_array[1:, :]) Y = np.float32(label_array[1:, :]) print("Image array shape: {0}".format(X.shape)) print("Label array shape: {0}".format(Y.shape)) end_time = cv2.getTickCount() print("Data set load duration: {0}" .format((end_time - start_time) // cv2.getTickFrequency())) return X, Y
def mk_ref(jmax,imax,f_list,tau,eref): ref=np.zeros(imax*jmax).reshape(jmax,imax) #t=cv2.getTickCount() for j in range(jmax): for i in range(imax): fref=f_list[j][i] temp=fref(tau[j,i],eref[j,i]) if temp < 0.0 : temp = 0.0 if temp > 1.0 : temp = 1.0 ref[j,i] = temp #if j % 100 == 0: print j,(cv2.getTickCount()-t)/cv2.getTickFrequency() print np.mean(ref),np.std(ref) return ref
def mk_tau(jmax,imax,f_list,eref,cref): taux=np.zeros(imax*jmax).reshape(jmax,imax) x=t_set #t=cv2.getTickCount() for j in range(jmax): for i in range(imax): fref=f_list[j][i] res=np.polyfit(x,fref(x,eref[j,i]),3) # 3rd order taux[j,i]=iestimate(res,cref[j,i]) #if j % 100 == 0: print i,(cv2.getTickCount()-t)/cv2.getTickFrequency() temp=np.where(np.isnan(taux)==True) print np.nanmean(taux),np.nanstd(taux),len(temp[0]) return taux #5/25/2016
def mk_rad(jmax,imax,inc,dem,sang,tau,ref,eref): rad=np.zeros(imax*jmax).reshape(jmax,imax) #t=cv2.getTickCount() for j in range(jmax): rad[j,:] = radiance(ref[j,:],inc[j,:],tau[j,:],dem[j,:]/1000,eref[j,:],sang[j,:]) #if j % 100 == 0: print j,(cv2.getTickCount()-t)/cv2.getTickFrequency() print np.mean(rad),np.std(rad) return rad
def __init__(self, haarCascadeFilePath, videoCapture): try: self.faceCascade = cv2.CascadeClassifier(haarCascadeFilePath) except: raise Exception("Error creating haar cascade classifier. Are you sure file " + haarCascadeFilePath + " exists?") self.videoCapture = videoCapture self.foundFace = False self.trackedFace = None self.trackedFaceROI = None self.templateMatchingStartTime = cv2.getTickCount() self.templateMatchingCurrentTime = cv2.getTickCount() self.isTemplateMatchingRunning = False self.img = None
def detectCascade(self, img, roiOnly=False): if roiOnly: searchArea = self.getSubRect(img, self.trackedFaceROI) else: searchArea = img width = searchArea.shape[0] faces = self.faceCascade.detectMultiScale(searchArea, scaleFactor = settings.cascadeScaleFactor, minNeighbors = settings.cascadeMinNeighbors, minSize = (int(width*settings.minimumFaceSize), int(width*settings.minimumFaceSize)), maxSize = (int(width*settings.maximumFaceSize), int(width*settings.maximumFaceSize))) if len(faces) == 0: if roiOnly and not self.isTemplateMatchingRunning: self.isTemplateMatchingRunning = True self.templateMatchingStartTime = cv2.getTickCount() elif not roiOnly: self.foundFace = False self.trackedFace = None; return self.foundFace=True # track only the largest face self.trackedFace = self.largestFace(faces) # adjust face position if necessary if roiOnly: self.trackedFace[0] += self.trackedFaceROI[0] self.trackedFace[1] += self.trackedFaceROI[1] self.trackedFaceTemplate = self.scaleRect(self.trackedFace, img, 0.5) self.trackedFaceROI = self.scaleRect(self.trackedFace, img, 2)
def main(): print os.getcwd() capture = cv2.VideoCapture('../data/road.avi') ticks = 0 img_prep = ImagePreparator() vis = Visualizer() while capture.isOpened(): prevTick = ticks ticks = cv2.getTickCount() t = (ticks - prevTick) / cv2.getTickFrequency() fps = int(1 / t) retval, image = capture.read() height, width, channels = image.shape rect = np.array([ [0, 200], [639, 200], [639, 359], [0, 359]], dtype="float32") dst = np.array([ [0, 0], [639, 0], [350, 699], [298, 699]], dtype="float32") # Aufbereitung des Bilder warped = img_prep.warp_perspective(image.copy(), rect, dst, (640, 700)) roi = img_prep.define_roi(warped, 0.6, 0, 0.40) gray = img_prep.grayscale(roi) blur = img_prep.blur(gray, (5, 5), 0) canny = img_prep.edge_detection(blur, 50, 150, 3) vis.draw_text(canny, 'FPS: ' + str(fps), 1, (255, 0, 0), (int(width*0.015), int(height*0.15))) vis.show(canny)
