Navigation

    VEYE IMAGING Forum

    • Register
    • Login
    • Search
    • Categories
    • Tags
    • Recent
    • Popular
    • Users
    • WIKI
    • veye.cc

    SOLVED RAW-MIPI-AR0234M配置为4lane模式

    Machine Vision camera
    2
    55
    7787
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • veye_xumm
      veye_xumm @veye_xumm last edited by veye_xumm

      1. 使用这个脚本,将camera的lane number配置为4,并保存参数。
      2. 修改dts源码,将2lane改为4lane。
        比如,对于jetson orin nano 8gb,可以修改dts源文件中的num_lanes为4。
      3. 编译dts源码,并部署实施。
      1 Reply Last reply Reply Quote 0
      • L
        lizi last edited by lizi

        您好,我想问一下这个60帧和120帧是基于raw格式下的吗?我用opencv去获取。就是我用opencv去获取frame,只能拿到640* 480分辨率下的60帧,如果设置为1920*1200就只有大概20帧。ar0234

        veye_xumm 1 Reply Last reply Reply Quote 0
        • veye_xumm
          veye_xumm @lizi last edited by

          @lizi
          34f8c1cd-f518-40eb-9c30-d1ea6f9dd309-image.png
          试一下这个命令,显示的帧率是多少?

          export WIDTH=1920
          export HEIGHT=1200
          export FPS=60
          v4l2-ctl --set-fmt-video=width=$WIDTH,height=$HEIGHT,pixelformat=GREY --stream-mmap --stream-count=-1 --stream-to=/dev/null
          
          L 1 Reply Last reply Reply Quote 0
          • L
            lizi @veye_xumm last edited by

            @veye_xumm said in RAW-MIPI-AR0234M配置为4lane模式:

            export WIDTH=1920
            export HEIGHT=1200
            export FPS=60
            v4l2-ctl --set-fmt-video=width=$WIDTH,height=$HEIGHT,pixelformat=GREY --stream-mmap --stream-count=-1 --stream-to=/dev/null

            60帧

            veye_xumm 1 Reply Last reply Reply Quote 0
            • veye_xumm
              veye_xumm @lizi last edited by

              @lizi 说明相机配置没问题,还是程序效率的问题。

              L 1 Reply Last reply Reply Quote 0
              • L
                lizi @veye_xumm last edited by lizi

                @veye_xumm
                截屏2024-01-19 17.45.12.png
                这是我的捕获代码,那就是opencv效率的问题吗

                veye_xumm L 2 Replies Last reply Reply Quote 0
                • veye_xumm
                  veye_xumm @lizi last edited by

                  @lizi opencv底层我不太懂。但是最底下肯定是跟我刚才发的v4l2命令类似的接口获取的v4l2 buffer,有没有做什么数据搬移,数据格式转换,这个我确定不了。
                  你有没有试过我们这个demo?

                  1 Reply Last reply Reply Quote 0
                  • L
                    lizi @lizi last edited by

                    刚刚在我的代码之前加了v4l2的cmd最后还是只有大概20张

                    veye_xumm 1 Reply Last reply Reply Quote 0
                    • veye_xumm
                      veye_xumm @lizi last edited by

                      @lizi 什么意思?

                      L 1 Reply Last reply Reply Quote 0
                      • L
                        lizi @veye_xumm last edited by lizi

                        @veye_xumm 就是把demo里最后循环里的cv2.imshow改成了计数。因为显示图片没法判断帧数。所以我改成了1s的计数。最后只有20截屏2024-01-19 17.58.13.png 截屏2024-01-19 17.58.33.png
                        把上面demo里的代码改成了下面的计数

                        veye_xumm 1 Reply Last reply Reply Quote 0
                        • veye_xumm
                          veye_xumm @lizi last edited by

                          @lizi 如此说来确实有这个问题。我们主要提供好驱动,opencv不是很熟。
                          你用jetson平台,最好还是用带jetson的硬件加速的开发包去开发。比如CUDA之类。
                          我们应用层和算法都不太熟悉。你可以到英伟达的开发者网站上找找资料。

                          L 1 Reply Last reply Reply Quote 0
                          • L
                            lizi @veye_xumm last edited by

                            @veye_xumm 那有没有办法用v4l2-ctl --set-fmt-video=width=$WIDTH,height=$HEIGHT,pixelformat=GREY --stream-mmap --stream-count=-1 --stream-to=/dev/null这个捕获帧的方法去捕获图片啊,这个获取的图片是raw格式的吗?

                            veye_xumm 1 Reply Last reply Reply Quote 0
                            • veye_xumm
                              veye_xumm @lizi last edited by

                              @lizi
                              我测了一下orin nano上面,实际采集数据的帧率是没有问题的。 你可以延长你的代码的统计时间到10秒,100秒,可以发现帧率会趋近于接近60fps。

                              那么为何出现你只统计1秒的时候,帧率只有20帧的结果呢? 我进行了一下简单的时间分析,发现是第一次调用cap.read()函数的时候,耗时大约半秒左右。(估计是底层进行buffer的初始化之类的操作。)

                              可以参考我贴上来这个代码:

                              import cv2
                              import argparse
                              import subprocess
                              import time
                              
                              def main():
                                  # Set up command-line argument parser
                                  parser = argparse.ArgumentParser(description='Real-time display of GREY image from /dev/video0')
                                  parser.add_argument('--roix', type=int, default=0, help='roi start x (default: 0)')
                                  parser.add_argument('--roiy', type=int, default=0, help='roi start y (default: 0)')
                                  parser.add_argument('--width', type=int, default=1920, help='image width (default: 640)')
                                  parser.add_argument('--height', type=int, default=1200, help='image height (default: 480)')
                                  parser.add_argument('--fps', type=int, default=60, help='frame rate (default: 30)')
                                  args = parser.parse_args()
                                  
                                  v4l2_cmd = f"v4l2-ctl --set-ctrl roi_x={args.roix}"
                                  subprocess.run(v4l2_cmd, shell=True)
                                  
                                  v4l2_cmd = f"v4l2-ctl --set-ctrl roi_y={args.roiy}"
                                  subprocess.run(v4l2_cmd, shell=True)
                                  
                                  v4l2_cmd = f"v4l2-ctl --set-fmt-video=width={args.width},height={args.height}"
                                  subprocess.run(v4l2_cmd, shell=True)
                                  
                                  v4l2_cmd = f"v4l2-ctl --set-ctrl frame_rate={args.fps}"
                                  subprocess.run(v4l2_cmd, shell=True)
                                  
                                  # Open the /dev/video0 device
                                  cap = cv2.VideoCapture('/dev/video0')
                                  if not cap.isOpened():
                                      print("Failed to open video device")
                                      return
                              
                                  # Set the image size
                                  cap.set(cv2.CAP_PROP_FRAME_WIDTH, args.width)
                                  cap.set(cv2.CAP_PROP_FRAME_HEIGHT, args.height)
                              
                                  frame_num = 0
                                  
                                  # Loop over frames and display them
                                  while True:
                                      # Read a frame
                                      ret, frame = cap.read()
                                      # Check if reading was successful
                                      if ret:
                                          if frame_num == 0:
                                              time_start = time.perf_counter()
                                          frame_num += 1
                                          # Display the frame
                                          # cv2.imshow('VEYE MV camera GREY image preview', frame)
                                          print(f"Success save , {frame_num}")
                                      else:
                                          print("Failed to read frame")
                                          #break
                              
                                      time_end = time.perf_counter()
                                      time_ms = (time_end - time_start)*1000
                                      print(time_ms)
                                      if time_ms > 1000:
                                          print(frame_num)
                                          break
                              
                                      # Exit if 'q' key is pressed
                                      if cv2.waitKey(1) & 0xFF == ord('q'):
                                          break
                              
                                  # Release resources
                                  cap.release()
                                  cv2.destroyAllWindows()
                              
                              if __name__ == '__main__':
                                  main()
                              
                              
                              L 1 Reply Last reply Reply Quote 0
                              • L
                                lizi @veye_xumm last edited by

                                @veye_xumm 好的

                                L 1 Reply Last reply Reply Quote 0
                                • L
                                  lizi @lizi last edited by lizi

                                  @veye_xumm 想问一下这里的曝光时间只有读取,没有写入的吗?设置曝光时间的命令是在哪里呢?在i2c.sh里截屏2024-01-23 12.10.07.png已经找到了

                                  veye_xumm 1 Reply Last reply Reply Quote 0
                                  • veye_xumm
                                    veye_xumm @lizi last edited by

                                    @lizi
                                    请参考一下这个链接的part4.1

                                    L 1 Reply Last reply Reply Quote 0
                                    • L
                                      lizi @veye_xumm last edited by

                                      This post is deleted!
                                      1 Reply Last reply Reply Quote 0
                                      • L
                                        lizi last edited by lizi

                                        @veye_xumm 截屏2024-01-23 16.22.05.png
                                        安装驱动出现这个错是为什么。板子是orin-nano 8gb ,然后mipi是ar0234

                                        veye_xumm 1 Reply Last reply Reply Quote 0
                                        • veye_xumm
                                          veye_xumm @lizi last edited by

                                          @lizi i2c不通,你查查硬件连接,比如fpc排线方向。

                                          L 1 Reply Last reply Reply Quote 0
                                          • L
                                            lizi @veye_xumm last edited by lizi

                                            @veye_xumm
                                            截屏2024-01-23 17.24.22.png
                                            这是两边的插线方向。这个是对的吗orin-nano的有数据的线朝外侧,mipi有数据的线朝下方
                                            截屏2024-01-23 17.04.52.png
                                            我参考这个来安装的

                                            L 1 Reply Last reply Reply Quote 0
                                            • First post
                                              Last post