使用受支持的摄像头无法在opencv中达到60fps

Question

对于我的项目，我需要处理高速率的帧，越多越好，但至少60帧是强制性的。我在Windows 10机器vía Anaconda中使用opencv 4.1.1，以防您需要了解平台。使用提供的代码，无论帧速率或分辨率如何，我都能获得30 fps

我的相机型号是ELP-USBFHD01M-L21，它能够：

1. 1920（高）x1080（V）MJPEG 30fps YUY2 6fps
2. 1280（高）x1024（高）兆焦耳每分钟30帧YUY2每分钟6帧
3. 1280（高）x 720（V）兆焦耳每小时60帧，比每小时2帧9帧
4. 1024（高）x 768（五）兆焦耳每小时30帧，比每小时2帧9帧
5. 800（高）x 600（V）MJPEG 60帧/秒YUY2 21帧/秒
6. 640（高）x480（V）MJPEG 120fps YUY2 30fps
7. 352（高）x288（V）MJPEG 120fps YUY2 30fps
8. 320（高）x240（伏）兆焦耳PEG 120fps YUY2 30fps

所有这些帧速率都使用AMCAP软件进行了正确测试

如何达到所需的帧速率

代码：

import cv2
import time
import numpy as np

if __name__ == "__main__":

    VIDEO_SOURCE = 0
    CAMERA_CAPTURE_FPS = 60
    CAMERA_CAPTURE_WIDTH = 1280
    CAMERA_CAPTURE_HEIGHT = 720

    frameCounter = 0
    fps = 0
    startTime = time.time()

    cap = cv2.VideoCapture(VIDEO_SOURCE)

    print("Parameters BEFORE assignment: ")
    print("WIDTH: " + str(cap.get(cv2.CAP_PROP_FRAME_WIDTH)))
    print("HEIGHT: " + str(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)))
    print("FPS: " + str(cap.get(cv2.CAP_PROP_FPS)))
    print("FOURCC: " + str(cap.get(cv2.CAP_PROP_FOURCC)))

    # Noting seems to change adding or deleting this 2 lines
    fourcc = cv2.VideoWriter_fourcc('M','J','P','G')
    cap.set(cv2.CAP_PROP_FOURCC, fourcc)
    
    cap.set(cv2.CAP_PROP_FRAME_WIDTH, CAMERA_CAPTURE_WIDTH)
    cap.set(cv2.CAP_PROP_FRAME_HEIGHT, CAMERA_CAPTURE_HEIGHT)
    cap.set(cv2.CAP_PROP_FPS, CAMERA_CAPTURE_FPS)

    print()
    print("Parameters AFTER assignment: ")
    print("WIDTH: " + str(cap.get(cv2.CAP_PROP_FRAME_WIDTH)))
    print("HEIGHT: " + str(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)))
    print("FPS: " + str(cap.get(cv2.CAP_PROP_FPS)))
    print("FOURCC: " + str(cap.get(cv2.CAP_PROP_FOURCC)))


    while True:

        ret, frame = cap.read()

        cv2.putText(frame, "FPS: " + str(fps), (20, 40), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 1, cv2.LINE_AA)
        frameCounter += 1

        if time.time() - startTime >= 1:
            fps = frameCounter
            frameCounter = 0
            startTime = time.time()

        cv2.imshow("Camera stream", frame)

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    cv2.destroyAllWindows()
    cap.release()