EVM.rar_closer7oc_evm_python 欧拉放大_socialj8x_欧拉视频放大

import cv2 import numpy as np import scipy.signal as signal import scipy.fftpack as fftpack import os def rgb2yiq(src): T = np.array([[0.299, 0.587,0.114 ], [ 0.596, -0.274,-0.322], [ 0.212, -0.523,0.312]]) dst = src.transpose(2,0,1) dst_re = dst.reshape((3,dst.shape[1]*dst.shape[2])) dst_re = np.dot(T,dst_re) dst_re = dst_re.reshape(3,dst.shape[1],dst.shape[2]) dst = dst_re.transpose(1,2,0) dst[dst<0] = 0 dst[dst>1] = 1 return dst def yiq2rgb(src): T = np.array([[1.0, 0.956, 0.620], [1.0, -0.272, -0.647],[1.0, -1.108, 1.705]]) dst = src.transpose(2,0,1) dst_re = dst.reshape((3,dst.shape[1]*dst.shape[2])) dst_re = np.dot(T,dst_re) dst_re = dst_re.reshape(3,dst.shape[1],dst.shape[2]) dst = dst_re.transpose(1,2,0) dst[dst<0] = 0 dst[dst>1] = 1 dst = np.uint8(dst*255) return dst #convert each of the frames to the YIQ color space. def convertVideo_YIQ(inputframes): # print(len(inputframes)) dstFrame=[] for frame in inputframes: #print(frame) img = rgb2yiq(frame) dstFrame.append(img) dstFrame = np.array(dstFrame) return dstFrame #convert each of the frames to the RGB color space. def convertVideo_BGR(inputframes): # print(len(inputframes)) dstFrame=[] for frame in inputframes: #print(frame) img = yiq2rgb(frame) img = cv2.cvtColor(img,cv2.COLOR_RGB2BGR) dstFrame.append(img) dstFrame = np.array(dstFrame) return dstFrame #Build Gaussian Pyramid def build_gaussian_pyramid(src,level=3): s=src.copy() pyramid=[s] for i in range(level): s=cv2.pyrDown(s) pyramid.append(s) return pyramid #Build Laplacian Pyramid ##def build_laplacian_pyramid(src,levels=3): ## gaussianPyramid = build_gaussian_pyramid(src, levels) ## pyramid=[] ## for i in range(levels,0,-1): ## GE=cv2.pyrUp(gaussianPyramid[i]) ## L=cv2.subtract(gaussianPyramid[i-1],GE) ## pyramid.append(L) ## return pyramid def build_laplacian_pyramid(src,levels=3): curr = src.copy() lapyramid=[] for i in range(levels-1): down = cv2.pyrDown(curr) up = cv2.pyrUp(down,curr.shape[:2]) Lap = curr - up Lap[Lap<0]=0 #L[L>1]=1 lapyramid.append(Lap) curr = down lapyramid.append(curr) return lapyramid[::-1] #load video from file def load_video(video_filename): cap=cv2.VideoCapture(video_filename) frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT)) width, height = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)),int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) fps = int(cap.get(cv2.CAP_PROP_FPS)) frames=np.zeros((frame_count,height,width,3),dtype='float') x=0 while cap.isOpened(): ret,frame=cap.read() if ret is True: frames[x]=frame x+=1 else: break return frames,fps # apply temporal ideal bandpass filter to gaussian video def temporal_ideal_filter(tensor,low,high,fps,axis=0): fft=fftpack.fft(tensor,axis=axis) frequencies = fftpack.fftfreq(tensor.shape[0], d=1.0 / fps) bound_low = (np.abs(frequencies - low)).argmin() bound_high = (np.abs(frequencies - high)).argmin() fft[:bound_low] = 0 fft[bound_high:-bound_high] = 0 fft[-bound_low:] = 0 iff=fftpack.ifft(fft, axis=axis) return np.abs(iff) # build gaussian pyramid for video def gaussian_video(video_tensor,levels=3): for i in range(0,video_tensor.shape[0]): frame=video_tensor[i] pyr=build_gaussian_pyramid(frame,level=levels) gaussian_frame=pyr[-1] if i==0: vid_data=np.zeros((video_tensor.shape[0],gaussian_frame.shape[0],gaussian_frame.shape[1],3)) vid_data[i]=gaussian_frame return vid_data #reconstract video from original video and gaussian video def reconstract_video(amp_video,origin_video,levels=3): final_video=np.zeros(origin_video.shape) for i in range(0,amp_video.shape[0]): img = amp_video[i] for x in range(levels): img=cv2.pyrUp(img) if levels > 3: img = cv2.resize(img,origin_video[i].shape[1::-1]) img=img+origin_video[i] final_video[i]=img return final_video #save video to files def save_video(save_video,outpath,name): os.makedirs(outpath,exist_ok=True) fourcc = cv2.VideoWriter_fourcc('M','J','P','G') [height,width]=save_video[0].shape[0:2] print(outpath+"/"+name+"_out.avi") writer = cv2.VideoWriter(outpath+"/"+name+"_out.avi", fourcc, 30, (width, height), 1) for i in range(0,save_video.shape[0]): writer.write(cv2.convertScaleAbs(save_video[i])) writer.release() #build laplacian pyramid for video def laplacian_video(video_tensor,levels=3): tensor_list=[] for i in range(0,video_tensor.shape[0]): frame=video_tensor[i] pyr=build_laplacian_pyramid(frame,levels=levels) if i==0: for k in range(levels): tensor_list.append(np.zeros((video_tensor.shape[0],pyr[k].shape[0],pyr[k].shape[1],3))) for n in range(levels): tensor_list[n][i] = pyr[n] return tensor_list #butterworth bandpass filter def butter_bandpass_filter(data, lowcut, highcut, fs, order=5): omega = 0.5 * fs low = lowcut / omega high = highcut / omega b, a = signal.butter(order, [low, high], btype='band') y = signal.lfilter(b, a, data, axis=0) return y #reconstract video from laplacian pyramid def reconstract_from_tensorlist(filter_tensor_list,levels=3): final=np.zeros(filter_tensor_list[-1].shape) for i in range(filter_tensor_list[0].shape[0]): up = filter_tensor_list[0][i] for n in range(levels-1): up=cv2.pyrUp(up) ## up=cv2.pyrUp(up)+filter_tensor_list[n + 1][i] final[i]=up return final #restore laplacian Image def restoe_LapImage(Lpyramid): print(len(Lpyramid)) levels=len(Lpyramid)-1 cur = Lpyramid[levels] for i in range(levels,0,-1): up = cv2.pyrUp(cur,Lpyramid[i-1].shape[:2]) cur = up + Lpyramid[i-1] return cur def loadvideo_getfps(videoName): frames=[] # os.makedirs(outpath,exist_ok=True) cap = cv2.VideoCapture("./data/"+videoName+".mp4") fps = cap.get(cv2.CAP_PROP_FPS) while cap.isOpened(): ret,frame=cap.read() if ret is True: frame = cv2.cvtColor(frame,cv2.COLOR_BGR2RGB) frame = np.double(frame) cv2.normalize(frame, frame, 1, 0, cv2.NORM_MINMAX) frames.append(frame) else: break frames = np.array(frames) return fps,frames #magnify color def magnify_color(video_name,low,high,levels=5,amplification=50): # rgbframes,fps = load_video(video_name) fps,rgbframes = loadvideo_getfps(video_name) print("rgb:",rgbframes.shape) yiqFrames = convertVideo_YIQ(rgbframes) print("yiq:",yiqFrames.shape) gau_video=gaussian_video(yiqFrames,levels=levels) print(gau_video.shape) filtered_tensor=temporal_ideal_filter(gau_video,low,high,fps) # amplified_video= filtered_tensor*amplification final=reconstract_video(amplified_video,yiqFrames,levels=levels) rgbfinal=convertVideo_BGR(final) save_video(rgbfinal,"color",video_name) #manify motion def magnify_motion(video_name,low,high,levels=3,amplification=50): fps,rgbframes = loadvideo_getfps(video_name) yiqFrames = convertVideo_YIQ(rgbframes) print("rgb:",rgbframes.shape,"yiq:",yiqFrames.shape) lap_video_list=laplacian_video(yiqFrames,levels=levels) filter_tensor_list=[] for i in range(lev