Last active
September 1, 2021 07:03
-
-
Save Asadullah-Dal17/fd71c31bac74ee84e6a31af50fa62961 to your computer and use it in GitHub Desktop.
Revisions
-
Asadullah-Dal17 revised this gist
Sep 1, 2021 . 1 changed file with 1 addition and 1 deletion.There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -86,7 +86,7 @@ def landmarksDetection(img, results, draw=False): frame =utils.textWithBackground(frame,f'FPS: {round(fps,1)}',FONTS, 1.0, (20, 50), bgOpacity=0.9, textThickness=2) # writing image for thumbnail drawing shape # cv.imwrite(f'img/frame_{frame_counter}.png', frame) cv.imshow('frame', frame) key = cv.waitKey(1) if key==ord('q') or key ==ord('Q'): -
Sep 1, 2021 . 1 changed file with 36 additions and 72 deletions.There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -1,17 +1,15 @@ import cv2 as cv import mediapipe as mp import time from numpy import greater import utils # variables frame_counter =0 # constants FONTS =cv.FONT_HERSHEY_COMPLEX # face bounder indices FACE_OVAL=[ 10, 338, 297, 332, 284, 251, 389, 356, 454, 323, 361, 288, 397, 365, 379, 378, 400, 377, 152, 148, 176, 149, 150, 136, 172, 58, 132, 93, 234, 127, 162, 21, 54, 103,67, 109] @@ -30,7 +28,7 @@ map_face_mesh = mp.solutions.face_mesh # camera object camera = cv.VideoCapture("VideoFile.mp4") # landmark detection function def landmarksDetection(img, results, draw=False): img_height, img_width= img.shape[:2] @@ -42,89 +40,55 @@ def landmarksDetection(img, results, draw=False): # returning the list of tuples for each landmarks return mesh_coord with map_face_mesh.FaceMesh(min_detection_confidence =0.5, min_tracking_confidence=0.5) as face_mesh: # starting time here start_time = time.time() # starting Video loop here. while True: frame_counter +=1 # frame counter ret, frame = camera.read() # getting frame from camera if not ret: break # no more frames break # resizing frame # frame = cv.resize(frame, None, fx=2.0, fy=2.0, interpolation=cv.INTER_CUBIC) # writing orginal image image thumbnail # cv.imwrite(f'img/img_{frame_counter}.png', frame) # print(frame_counter) rgb_frame = cv.cvtColor(frame, cv.COLOR_RGB2BGR) results = face_mesh.process(rgb_frame) if results.multi_face_landmarks: mesh_coords = landmarksDetection(frame, results, False) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in FACE_OVAL], utils.WHITE, opacity=0.4) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in LEFT_EYE], utils.GREEN, opacity=0.4) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in RIGHT_EYE], utils.GREEN, opacity=0.4) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in LEFT_EYEBROW], utils.ORANGE, opacity=0.4) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in RIGHT_EYEBROW], utils.ORANGE, opacity=0.4) frame =utils.fillPolyTrans(frame, [mesh_coords[p] for p in LIPS], utils.BLACK, opacity=0.3 ) # Changes for Thumbnail of youtube Video [cv.circle(frame,mesh_coords[p], 2, utils.GREEN , -1) for p in LIPS] [cv.circle(frame,mesh_coords[p], 2, utils.BLACK ,- 1, cv.LINE_AA) for p in RIGHT_EYE] [cv.circle(frame,mesh_coords[p], 2, utils.BLACK , -1) for p in LEFT_EYE] [cv.circle(frame,mesh_coords[p], 2, utils.BLACK , -1) for p in RIGHT_EYEBROW] [cv.circle(frame,mesh_coords[p], 2, utils.BLACK , -1) for p in LEFT_EYEBROW] [cv.circle(frame,mesh_coords[p], 2, utils.RED , -1) for p in FACE_OVAL] # calculating frame per seconds FPS end_time = time.time()-start_time fps = frame_counter/end_time frame =utils.textWithBackground(frame,f'FPS: {round(fps,1)}',FONTS, 1.0, (20, 50), bgOpacity=0.9, textThickness=2) # writing image for thumbnail drawing shape cv.imwrite(f'img/frame_{frame_counter}.png', frame) cv.imshow('frame', frame) key = cv.waitKey(1) if key==ord('q') or key ==ord('Q'): break cv.destroyAllWindows() -
Sep 1, 2021 . 1 changed file with 126 additions and 145 deletions.There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -1,150 +1,131 @@ import cv2 as cv import mediapipe as mp import time import utils import math # variables FRAME_COUNTER =0 TOTAL_BLINKS = 0 CLOSED_EYES_FRAME_COUNTER =0 # Counts frames while Eyes are closed # constants FONTS =cv.FONT_HERSHEY_COMPLEX CLOSED_EYES_FRAME = 3 # threshold to count a blink, after number of frames # face bounder indices FACE_OVAL=[ 10, 338, 297, 332, 284, 251, 389, 356, 454, 323, 361, 288, 397, 365, 379, 378, 400, 377, 152, 148, 176, 149, 150, 136, 172, 58, 132, 93, 234, 127, 162, 21, 54, 103,67, 109] # lips indices for Landmarks LIPS=[ 61, 146, 91, 181, 84, 17, 314, 405, 321, 375,291, 308, 324, 318, 402, 317, 14, 87, 178, 88, 95,185, 40, 39, 37,0 ,267 ,269 ,270 ,409, 415, 310, 311, 312, 13, 82, 81, 42, 183, 78 ] LOWER_LIPS =[61, 146, 91, 181, 84, 17, 314, 405, 321, 375, 291, 308, 324, 318, 402, 317, 14, 87, 178, 88, 95] UPPER_LIPS=[ 185, 40, 39, 37,0 ,267 ,269 ,270 ,409, 415, 310, 311, 312, 13, 82, 81, 42, 183, 78] # Left eyes indices LEFT_EYE =[ 362, 382, 381, 380, 374, 373, 390, 249, 263, 466, 388, 387, 386, 385,384, 398 ] LEFT_EYEBROW =[ 336, 296, 334, 293, 300, 276, 283, 282, 295, 285 ] # right eyes indices RIGHT_EYE=[ 33, 7, 163, 144, 145, 153, 154, 155, 133, 173, 157, 158, 159, 160, 161 , 246 ] RIGHT_EYEBROW=[ 70, 63, 105, 66, 107, 55, 65, 52, 53, 46 ] map_face_mesh = mp.solutions.face_mesh # camera object camera = cv.VideoCapture(0) # landmark detection function def landmarksDetection(img, results, draw=False): img_height, img_width= img.shape[:2] # list[(x,y), (x,y)....] mesh_coord = [(int(point.x * img_width), int(point.y * img_height)) for point in results.multi_face_landmarks[0].landmark] if draw : [cv.circle(img, p, 2, utils.GREEN, -1) for p in mesh_coord] # returning the list of tuples for each landmarks return mesh_coord # finding distance between two points def euclideanDistance(point1, point2): x, y = point1 x1, y1 = point2 euclidean_dist = math.sqrt((x1 - x) ** 2 + (y1 - y) ** 2) return euclidean_dist # Blinking Detector Function def blinkRatio(frame, landmarks, right_eye_indics, left_eye_indics): # getting the points # Right Eye horizontal points rh_right = landmarks[right_eye_indics[0]] rh_left = landmarks[right_eye_indics[7]] # left Eyes horizontal points lh_right = landmarks[left_eye_indics[0]] lh_left = landmarks[left_eye_indics[7]] cv.circle(frame, rh_right, 2, utils.GREEN, -1) cv.circle(frame, rh_left, 2, utils.WHITE, -1) cv.circle(frame, lh_right, 2, utils.GREEN, -1) cv.circle(frame, lh_left, 2, utils.WHITE, -1) # vertical points for Right Eyes rv_top= landmarks[right_eye_indics[12]] rv_bottom = landmarks[right_eye_indics[4]] # Vertical points for Left Eye lv_top = landmarks[left_eye_indics[12]] lv_bottom = landmarks[left_eye_indics[4]] cv.circle(frame, rv_top, 2, utils.GREEN, -1) cv.circle(frame, rv_bottom, 2, utils.WHITE, -1) cv.circle(frame, lv_top, 2, utils.GREEN, -1) cv.circle(frame, lv_bottom, 2, utils.WHITE, -1) # finding euclidean distance h_right_euclid_dist =euclideanDistance(rh_right, rh_left) v_right_euclid_dist =euclideanDistance(rv_top, rv_bottom) h_left_euclid_dist = euclideanDistance(lh_right, lh_left) v_left_euclid_dist = euclideanDistance(lv_top, lv_bottom) eyes_ratio =( (h_right_euclid_dist/v_right_euclid_dist) + (h_left_euclid_dist/ v_left_euclid_dist))/2 return eyes_ratio with map_face_mesh.FaceMesh(min_detection_confidence =0.5, min_tracking_confidence=0.5) as face_mesh: # starting time here start_time = time.time() # starting Video loop here. while True: FRAME_COUNTER +=1 # frame counter ret, frame = camera.read() # getting frame from camera fr_height, fr_width = frame.shape[:2] if not ret: break # no more frames break # frame = cv.resize(frame, None, fx=2, fy=2, interpolation=cv.INTER_AREA) rgb_frame = cv.cvtColor(frame, cv.COLOR_RGB2BGR) results = face_mesh.process(rgb_frame) if results.multi_face_landmarks: mesh_coords = landmarksDetection(frame, results, False) # blinking ratio ratio = blinkRatio(frame, mesh_coords, RIGHT_EYE, LEFT_EYE) if ratio >4.1: CLOSED_EYES_FRAME_COUNTER +=1 print(CLOSED_EYES_FRAME_COUNTER, ratio) cv.putText(frame, f'Blink', (int(fr_width/2), 30), FONTS, 1.4, utils.GREEN, 2, cv.LINE_AA) else: if CLOSED_EYES_FRAME_COUNTER>CLOSED_EYES_FRAME: TOTAL_BLINKS +=1 CLOSED_EYES_FRAME_COUNTER=0 cv.putText(frame, f'Total Blinks: {TOTAL_BLINKS}', (int(20), 100), FONTS, 0.5, utils.GREEN, 1, cv.LINE_AA) # calculating frame per seconds FPS end_time = time.time()-start_time fps = FRAME_COUNTER/end_time frame =utils.textWithBackground(frame,f'FPS: {round(fps,1)}',FONTS, 1.0, (20, 50), bgOpacity=0.9, textThickness=2) cv.imshow('frame', frame) key = cv.waitKey(2) if key==ord('q') or key ==ord('Q'): break cv.destroyAllWindows() camera.release() -
Sep 1, 2021 . 1 changed file with 141 additions and 87 deletions.There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -4,93 +4,147 @@ Youtube Channel: https://www.youtube.com/c/aiphile ''' import cv2 as cv import numpy as np # colors # values =(blue, green, red) opencv accepts BGR values not RGB BLACK = (0,0,0) WHITE = (255,255,255) BLUE = (255,0,0) RED = (0,0,255) CYAN = (255,255,0) YELLOW =(0,255,255) MAGENTA = (255,0,255) GRAY = (128,128,128) GREEN = (0,255,0) PURPLE = (128,0,128) ORANGE = (0,165,255) PINK = (147,20,255) points_list =[(200, 300), (150, 150), (400, 200)] def drawColor(img, colors): x, y = 0,10 w, h = 20, 30 for color in colors: x += w+5 # y += 10 cv.rectangle(img, (x-6, y-5 ), (x+w+5, y+h+5), (10, 50, 10), -1) cv.rectangle(img, (x, y ), (x+w, y+h), color, -1) def textWithBackground(img, text, font, fontScale, textPos, textThickness=1,textColor=(0,255,0), bgColor=(0,0,0), pad_x=3, pad_y=3, bgOpacity=0.5): """ Draws text with background, with control transparency @param img:(mat) which you want to draw text @param text: (string) text you want draw @param font: fonts face, like FONT_HERSHEY_COMPLEX, FONT_HERSHEY_PLAIN etc. @param fontScale: (double) the size of text, how big it should be. @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be. @param textColor: tuple(BGR), values -->0 to 255 each @param bgColor: tuple(BGR), values -->0 to 255 each @param pad_x: int(pixels) padding of in x direction @param pad_y: int(pixels) 1 to 1.0 (), controls transparency of text background @return: img(mat) with draw with background """ (t_w, t_h), _= cv.getTextSize(text, font, fontScale, textThickness) # getting the text size x, y = textPos overlay = img.copy() # coping the image cv.rectangle(overlay, (x-pad_x, y+ pad_y), (x+t_w+pad_x, y-t_h-pad_y), bgColor,-1) # draw rectangle new_img = cv.addWeighted(overlay, bgOpacity, img, 1 - bgOpacity, 0) # overlaying the rectangle on the image. cv.putText(new_img,text, textPos,font, fontScale, textColor,textThickness ) # draw in text img = new_img return img def textBlurBackground(img, text, font, fontScale, textPos, textThickness=1,textColor=(0,255,0),kneral=(33,33) , pad_x=3, pad_y=3): """ Draw text with background blured, control the blur value, with kernal(odd, odd) @param img:(mat) which you want to draw text @param text: (string) text you want draw @param font: fonts face, like FONT_HERSHEY_COMPLEX, FONT_HERSHEY_PLAIN etc. @param fontScale: (double) the size of text, how big it should be. @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be. @param textColor: tuple(BGR), values -->0 to 255 each @param kneral: tuple(3,3) int as odd number: higher the value, more blurry background would be @param pad_x: int(pixels) padding of in x direction @param pad_y: int(pixels) padding of in y direction @return: img mat, with text drawn, with background blured call the function: img =textBlurBackground(img, 'Blured Background Text', cv2.FONT_HERSHEY_COMPLEX, 0.9, (20, 60),2, (0,255, 0), (49,49), 13, 13 ) """ (t_w, t_h), _= cv.getTextSize(text, font, fontScale, textThickness) # getting the text size x, y = textPos blur_roi = img[y-pad_y-t_h: y+pad_y, x-pad_x:x+t_w+pad_x] # croping Text Background img[y-pad_y-t_h: y+pad_y, x-pad_x:x+t_w+pad_x]=cv.blur(blur_roi, kneral) # merging the blured background to img cv.putText(img,text, textPos,font, fontScale, textColor,textThickness ) # cv.imshow('blur roi', blur_roi) # cv.imshow('blured', img) return img def fillPolyTrans(img, points, color, opacity): """ @param img: (mat) input image, where shape is drawn. @param points: list [tuples(int, int) these are the points custom shape,FillPoly @param color: (tuples (int, int, int) @param opacity: it is transparency of image. @return: img(mat) image with rectangle draw. """ list_to_np_array = np.array(points, dtype=np.int32) overlay = img.copy() # coping the image cv.fillPoly(overlay,[list_to_np_array], color ) new_img = cv.addWeighted(overlay, opacity, img, 1 - opacity, 0) # print(points_list) img = new_img return img def rectTrans(img, pt1, pt2, color, thickness, opacity): """ @param img: (mat) input image, where shape is drawn. @param pt1: tuple(int,int) it specifies the starting point(x,y) os rectangle @param pt2: tuple(int,int) it nothing but width and height of rectangle @param color: (tuples (int, int, int), it tuples of BGR values @param thickness: it thickness of board line rectangle, if (-1) passed then rectangle will be fulled with color. @param opacity: it is transparency of image. @return: """ overlay = img.copy() cv.rectangle(overlay, pt1, pt2, color, thickness) new_img = cv.addWeighted(overlay, opacity, img, 1 - opacity, 0) # overlaying the rectangle on the image. img = new_img return img def main(): cap = cv.VideoCapture('Girl.mp4') counter =0 while True: success, img = cap.read() # img = np.zeros((1000,1000, 3), dtype=np.uint8) img=rectTrans(img, pt1=(30, 320), pt2=(160, 260), color=(0,255,255),thickness=-1, opacity=0.6) img =fillPolyTrans(img=img, points=points_list, color=(0,255,0), opacity=.5) drawColor(img, [BLACK,WHITE ,BLUE,RED,CYAN,YELLOW,MAGENTA,GRAY ,GREEN,PURPLE,ORANGE,PINK]) textBlurBackground(img, 'Blured Background Text', cv.FONT_HERSHEY_COMPLEX, 0.8, (60, 140),2, YELLOW, (71,71), 13, 13) img=textWithBackground(img, 'Colored Background Texts', cv.FONT_HERSHEY_SIMPLEX, 0.8, (60,80), textThickness=2, bgColor=GREEN, textColor=BLACK, bgOpacity=0.7, pad_x=6, pad_y=6) imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY) # cv.imwrite('color_image.png', img) counter +=1 cv.imshow('img', img) cv.imwrite(f'image/image_{counter}.png', img) if cv.waitKey(1) ==ord('q'): break if __name__ == "__main__": main() -
Aug 18, 2021 . 1 changed file with 0 additions and 150 deletions.There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -1,150 +0,0 @@ -
Aug 18, 2021 .There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -0,0 +1,96 @@ ''' Author: Asadullah Dal Youtube Channel: https://www.youtube.com/c/aiphile ''' import cv2 as cv import numpy as np import mediapipe as mp import utils import time # Landmarks indices for different parts of face, mediapipe. # face bounder indices FACE_OVAL=[ 10, 338, 297, 332, 284, 251, 389, 356, 454, 323, 361, 288, 397, 365, 379, 378, 400, 377, 152, 148, 176, 149, 150, 136, 172, 58, 132, 93, 234, 127, 162, 21, 54, 103,67, 109] # lips indices for Landmarks LIPS=[ 61, 146, 91, 181, 181, 84, 17, 314, 405, 321, 375, 61, 185, 40, 39, 37,0 ,267 ,269 ,270 ,409, 78, 95 ,88 ,178 ,87, 14, 317, 402, 318, 324, 78, 191, 80, 81, 82, 13, 312, 311, 310, 415 ] # Left eyes indices LEFT_EYE =[ 362, 382, 381, 380, 374, 373, 390, 249, 263, 466, 388, 387, 386, 385,384, 398 ] LEFT_EYEBROW =[ 336, 296, 334, 293, 300, 276, 283, 282, 295, 285 ] # right eyes indices RIGHT_EYE=[ 33, 7, 163, 144, 145, 153, 154, 155, 133, 173, 157, 158, 159, 160, 161 , 246 ] RIGHT_EYEBROW=[ 70, 63, 105, 66, 107, 55, 65, 52, 53, 46 ] # Variables # Frame per seconds frame_counter =0 # Setting up mediapipe map_face_mesh= mp.solutions.face_mesh # Different function def faceLandmarksDetector(img, result, draw=False): # image width and height img_height, img_width = img.shape[:2] # getting all the landmark normalized cordinate(x,y) in the image # multiplying these coordinate with width and height we get image coordinates mesh_cord_point = [ (int(p.x*img_width) , int(p.y*img_height)) for p in results.multi_face_landmarks[0].landmark] if draw : [cv.circle(img, point_cord, 2, utils.GREEN, -1) for point_cord in mesh_cord_point] # print() return mesh_cord_point # setting up camera cap = cv.VideoCapture(0) # configring mediapipe for face mesh detection with map_face_mesh.FaceMesh( min_detection_confidence=0.5, min_tracking_confidence=0.5 ) as face_mesh: # string video/webcame feed here # initial time set here starting_time =time.time() while True: ret, frame = cap.read() frame_counter +=1 # converting color space from BGR to RGB rgb_frame = cv.cvtColor(frame, cv.COLOR_BGR2RGB) # getting the frame width height, width = frame.shape[:2] # getting the landmarks results = face_mesh.process(rgb_frame) # checking if landmarks are detected or not if results.multi_face_landmarks: # calling faceLandmarksDetector function and getting coordinate of each point in face mesh mesh_cords =faceLandmarksDetector(img=frame, result=results) frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in FACE_OVAL],color=utils.BLACK, opacity=0.4) # draw lips landmarks portion frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in LIPS],color=utils.WHITE, opacity=0.4) frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in LEFT_EYE],color=utils.YELLOW, opacity=0.4) frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in LEFT_EYEBROW],color=utils.GREEN, opacity=0.4) frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in RIGHT_EYE],color=utils.YELLOW, opacity=0.4) frame = utils.fillPolyTrans(img=frame,points=[mesh_cords[p] for p in RIGHT_EYEBROW],color=utils.GREEN, opacity=0.4) # calculating the end time of total frames end_time = time.time()- starting_time # calculating the frames per seconds fps = frame_counter/end_time cv.imshow('frame',frame) key = cv.waitKey(1) if key==ord('q'): break cv.destroyAllWindows() cap.release() This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -0,0 +1,150 @@ ''' Author: Asadullah Dal Youtube Channel: https://www.youtube.com/c/aiphile ''' import cv2 as cv import numpy as np # colors # values =(blue, green, red) opencv accepts BGR values not RGB BLACK = (0,0,0) WHITE = (255,255,255) BLUE = (255,0,0) RED = (0,0,255) CYAN = (255,255,0) YELLOW =(0,255,255) MAGENTA = (255,0,255) GRAY = (128,128,128) GREEN = (0,255,0) PURPLE = (128,0,128) ORANGE = (0,165,255) PINK = (147,20,255) points_list =[(200, 300), (150, 150), (400, 200)] def drawColor(img, colors): x, y = 0,10 w, h = 20, 30 for color in colors: x += w+5 # y += 10 cv.rectangle(img, (x-6, y-5 ), (x+w+5, y+h+5), (10, 50, 10), -1) cv.rectangle(img, (x, y ), (x+w, y+h), color, -1) def textWithBackground(img, text, font, fontScale, textPos, textThickness=1,textColor=(0,255,0), bgColor=(0,0,0), pad_x=3, pad_y=3, bgOpacity=0.5): """ Draws text with background, with control transparency @param img:(mat) which you want to draw text @param text: (string) text you want draw @param font: fonts face, like FONT_HERSHEY_COMPLEX, FONT_HERSHEY_PLAIN etc. @param fontScale: (double) the size of text, how big it should be. @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be. @param textColor: tuple(BGR), values -->0 to 255 each @param bgColor: tuple(BGR), values -->0 to 255 each @param pad_x: int(pixels) padding of in x direction @param pad_y: int(pixels) 1 to 1.0 (), controls transparency of text background @return: img(mat) with draw with background """ (t_w, t_h), _= cv.getTextSize(text, font, fontScale, textThickness) # getting the text size x, y = textPos overlay = img.copy() # coping the image cv.rectangle(overlay, (x-pad_x, y+ pad_y), (x+t_w+pad_x, y-t_h-pad_y), bgColor,-1) # draw rectangle new_img = cv.addWeighted(overlay, bgOpacity, img, 1 - bgOpacity, 0) # overlaying the rectangle on the image. cv.putText(new_img,text, textPos,font, fontScale, textColor,textThickness ) # draw in text img = new_img return img def textBlurBackground(img, text, font, fontScale, textPos, textThickness=1,textColor=(0,255,0),kneral=(33,33) , pad_x=3, pad_y=3): """ Draw text with background blured, control the blur value, with kernal(odd, odd) @param img:(mat) which you want to draw text @param text: (string) text you want draw @param font: fonts face, like FONT_HERSHEY_COMPLEX, FONT_HERSHEY_PLAIN etc. @param fontScale: (double) the size of text, how big it should be. @param textPos: tuple(x,y) position where you want to draw text @param textThickness:(int) fonts weight, how bold it should be. @param textColor: tuple(BGR), values -->0 to 255 each @param kneral: tuple(3,3) int as odd number: higher the value, more blurry background would be @param pad_x: int(pixels) padding of in x direction @param pad_y: int(pixels) padding of in y direction @return: img mat, with text drawn, with background blured call the function: img =textBlurBackground(img, 'Blured Background Text', cv2.FONT_HERSHEY_COMPLEX, 0.9, (20, 60),2, (0,255, 0), (49,49), 13, 13 ) """ (t_w, t_h), _= cv.getTextSize(text, font, fontScale, textThickness) # getting the text size x, y = textPos blur_roi = img[y-pad_y-t_h: y+pad_y, x-pad_x:x+t_w+pad_x] # croping Text Background img[y-pad_y-t_h: y+pad_y, x-pad_x:x+t_w+pad_x]=cv.blur(blur_roi, kneral) # merging the blured background to img cv.putText(img,text, textPos,font, fontScale, textColor,textThickness ) # cv.imshow('blur roi', blur_roi) # cv.imshow('blured', img) return img def fillPolyTrans(img, points, color, opacity): """ @param img: (mat) input image, where shape is drawn. @param points: list [tuples(int, int) these are the points custom shape,FillPoly @param color: (tuples (int, int, int) @param opacity: it is transparency of image. @return: img(mat) image with rectangle draw. """ list_to_np_array = np.array(points, dtype=np.int32) overlay = img.copy() # coping the image cv.fillPoly(overlay,[list_to_np_array], color ) new_img = cv.addWeighted(overlay, opacity, img, 1 - opacity, 0) # print(points_list) img = new_img return img def rectTrans(img, pt1, pt2, color, thickness, opacity): """ @param img: (mat) input image, where shape is drawn. @param pt1: tuple(int,int) it specifies the starting point(x,y) os rectangle @param pt2: tuple(int,int) it nothing but width and height of rectangle @param color: (tuples (int, int, int), it tuples of BGR values @param thickness: it thickness of board line rectangle, if (-1) passed then rectangle will be fulled with color. @param opacity: it is transparency of image. @return: """ overlay = img.copy() cv.rectangle(overlay, pt1, pt2, color, thickness) new_img = cv.addWeighted(overlay, opacity, img, 1 - opacity, 0) # overlaying the rectangle on the image. img = new_img return img def main(): cap = cv.VideoCapture('Girl.mp4') counter =0 while True: success, img = cap.read() # img = np.zeros((1000,1000, 3), dtype=np.uint8) img=rectTrans(img, pt1=(30, 320), pt2=(160, 260), color=(0,255,255),thickness=-1, opacity=0.6) img =fillPolyTrans(img=img, points=points_list, color=(0,255,0), opacity=.5) drawColor(img, [BLACK,WHITE ,BLUE,RED,CYAN,YELLOW,MAGENTA,GRAY ,GREEN,PURPLE,ORANGE,PINK]) textBlurBackground(img, 'Blured Background Text', cv.FONT_HERSHEY_COMPLEX, 0.8, (60, 140),2, YELLOW, (71,71), 13, 13) img=textWithBackground(img, 'Colored Background Texts', cv.FONT_HERSHEY_SIMPLEX, 0.8, (60,80), textThickness=2, bgColor=GREEN, textColor=BLACK, bgOpacity=0.7, pad_x=6, pad_y=6) imgGray = cv.cvtColor(img, cv.COLOR_BGR2GRAY) # cv.imwrite('color_image.png', img) counter +=1 cv.imshow('img', img) cv.imwrite(f'image/image_{counter}.png', img) if cv.waitKey(1) ==ord('q'): break if __name__ == "__main__": main()