Recherche avancée

Sur d’autres sites (9649)

• ffmpeg - detect frames that have a given image/logo

  13 juillet 2012, par sofia

  I'm trying to split a video by detecting the presence of a marker (an image) in the frames. I've gone over the documentation and I see removelogo but not detectlogo ;)

  Does anyone know how this could be achieved ? I know what the logo is and the region it will be on.

  I'm thinking i can extract all frames to png's and then analyse them one by one (or n by n) but it might be a lengthy process..

  Any pointers ?

  Thanks :)

• Anomalie #4543 : Accessibilité des chargements ajax (live regions)

  4 septembre 2020, par nicod _

  Si sur le site de cette région, ya pas mal d’ajax qui se recharge "tout seul" sans interaction, possible oui ça va pas du tout,

  Non non, ce sont les cas classiques d’utilisation "de base" : formulaires, et listes de résultats avec paginations.

• OpenCV - motion detection in large video files

  18 décembre 2023, par M9A

  I have large video files (over a few GB each) that I am trying detect motion on. I am using fairly basic OpenCV code in python with ROI to detect motion in a certain area of the video only. This works absolutely fine but takes absolutely ages to process large files. Is there any way to speed things up ? I am simply just checking to see which file has motion detected in it and will not be outputting a video. The moment the first motion is detected the rest of the file is not checked.

  


  import sys
import cv2

video_path = “video.mp4”

capture = cv2.VideoCapture(video_path)

# Set the first frame as the reference background
_, background = capture.read()

# Define the region of interest (ROI)
x, y, w, h = 1468, 1142, 412, 385
roi = (x, y, x + w, y + h)

while capture.isOpened():
    ret, frame = capture.read()

    if not ret:
        break

    # Extract the region of interest from the current frame
    frame_roi = frame[y:y + h, x:x + w]

    # Calculate the absolute difference between the ROI and the background
    diff = cv2.absdiff(background[y:y + h, x:x + w], frame_roi)
    gray = cv2.cvtColor(diff, cv2.COLOR_BGR2GRAY)
    _, thresh = cv2.threshold(gray, 30, 255, cv2.THRESH_BINARY)

    # Check if motion is detected in the ROI (adjust threshold as needed)
    if cv2.countNonZero(thresh) > 75:
        print("YES")
        break

# Release the video capture object
capture.release()