【树莓派5 测评】 + 11.USB-Camera功能测试（zmj）

赵美京?????

【树莓派5 测评】 + 11.USB-Camera功能测试（zmj）

本篇主要介绍在树莓派5开发板测试USB-Camera（UVC）的流程。

//------UVC / USB-Camera
UVC通俗的讲就是USB摄像头，全称为USB Video Class，即：USB视频类，是一种为USB视频捕获设备定义的协议标准。

---------------------------以下为正文---------------------------

1.        摄像头接口和设备

摄像头设备位于” /dev/video*“（此处仅连接了USB摄像头）。

1. //------查看摄像头设备节点
   
2. //---测试指令
   
3. a.通过设备列表查看
   
4. ls -la /dev/video*
   
5. b.通过v4l2-ctl工具查看
   
6. v4l2-ctl --list-devices
   
7. c.通过media0检查有效摄像头节点
   
8. media-ctl -d /dev/media3 -p
   
9. d.无效指令：libcamera-hello无法检测UVC，貌似支持CSI接口摄像头
   
10. libcamera-hello --list-cameras
    
11. //---检查结果
    
12. USB 2.0 Camera: USB Camera (usb-xhci-hcd.0-2.1):
    
13.         /dev/video0
    
14.         /dev/video1
    
15.         /dev/media3
    
16.         
    
17. //------示例Log信息
    
18. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ lsusb
    
19. Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    
20. Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    
21. Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    
22. Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    
23. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ ls -la /dev/video*
    
24. crw-rw----+ 1 root video 81, 18 Apr 26 18:15 /dev/video19
    
25. crw-rw----+ 1 root video 81,  0 Apr 26 18:15 /dev/video20
    
26. crw-rw----+ 1 root video 81,  1 Apr 26 18:15 /dev/video21
    
27. crw-rw----+ 1 root video 81,  2 Apr 26 18:15 /dev/video22
    
28. crw-rw----+ 1 root video 81,  3 Apr 26 18:15 /dev/video23
    
29. crw-rw----+ 1 root video 81,  4 Apr 26 18:15 /dev/video24
    
30. crw-rw----+ 1 root video 81,  5 Apr 26 18:15 /dev/video25
    
31. crw-rw----+ 1 root video 81,  6 Apr 26 18:15 /dev/video26
    
32. crw-rw----+ 1 root video 81,  7 Apr 26 18:15 /dev/video27
    
33. crw-rw----+ 1 root video 81,  8 Apr 26 18:15 /dev/video28
    
34. crw-rw----+ 1 root video 81,  9 Apr 26 18:15 /dev/video29
    
35. crw-rw----+ 1 root video 81, 10 Apr 26 18:15 /dev/video30
    
36. crw-rw----+ 1 root video 81, 11 Apr 26 18:15 /dev/video31
    
37. crw-rw----+ 1 root video 81, 12 Apr 26 18:15 /dev/video32
    
38. crw-rw----+ 1 root video 81, 13 Apr 26 18:15 /dev/video33
    
39. crw-rw----+ 1 root video 81, 14 Apr 26 18:15 /dev/video34
    
40. crw-rw----+ 1 root video 81, 15 Apr 26 18:15 /dev/video35
    
41. crw-rw----+ 1 root video 81, 16 Apr 26 18:15 /dev/video36
    
42. crw-rw----+ 1 root video 81, 17 Apr 26 18:15 /dev/video37
    
43. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ lsusb
    
44. Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    
45. Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    
46. Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    
47. Bus 001 Device 004: ID 1b3f:2008 Generalplus Technology Inc. Usb Audio Device
    
48. Bus 001 Device 003: ID 0c45:6362 Microdia USB 2.0 Camera
    
49. Bus 001 Device 002: ID 1a40:0801 Terminus Technology Inc. USB 2.0 Hub
    
50. Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    
51. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ dmesg | tail -23
    
52. [148218.886321] usb 1-2: new high-speed USB device number 2 using xhci-hcd
    
53. [148219.034490] usb 1-2: New USB device found, idVendor=1a40, idProduct=0801, bcdDevice= 1.00
    
54. [148219.034496] usb 1-2: New USB device strings: Mfr=0, Product=1, SerialNumber=0
    
55. [148219.034499] usb 1-2: Product: USB 2.0 Hub
    
56. [148219.082777] hub 1-2:1.0: USB hub found
    
57. [148219.082819] hub 1-2:1.0: 4 ports detected
    
58. [148219.370336] usb 1-2.1: new high-speed USB device number 3 using xhci-hcd
    
59. [148219.548936] usb 1-2.1: New USB device found, idVendor=0c45, idProduct=6362, bcdDevice= 0.00
    
60. [148219.548944] usb 1-2.1: New USB device strings: Mfr=2, Product=1, SerialNumber=0
    
61. [148219.548947] usb 1-2.1: Product: USB 2.0 Camera
    
62. [148219.548949] usb 1-2.1: Manufacturer: Sonix Technology Co., Ltd.
    
63. [148219.694326] usb 1-2.3: new full-speed USB device number 4 using xhci-hcd
    
64. [148219.815720] usb 1-2.3: config 1 has an invalid interface number: 3 but max is 2
    
65. [148219.815725] usb 1-2.3: config 1 has no interface number 1
    
66. [148219.816405] usb 1-2.3: New USB device found, idVendor=1b3f, idProduct=2008, bcdDevice= 1.00
    
67. [148219.816409] usb 1-2.3: New USB device strings: Mfr=1, Product=2, SerialNumber=0
    
68. [148219.816412] usb 1-2.3: Product: Usb Audio Device
    
69. [148219.816415] usb 1-2.3: Manufacturer: Generalplus
    
70. [148219.851468] input: Generalplus Usb Audio Device as /devices/platform/axi/1000120000.pcie/1f00200000.usb/xhci-hcd.0/usb1/1-2/1-2.3/1-2.3:1.3/0003:1B3F:2008.0001/input/input5
    
71. [148219.910418] hid-generic 0003:1B3F:2008.0001: input,hidraw0: USB HID v2.01 Device [Generalplus Usb Audio Device] on usb-xhci-hcd.0-2.3/input3
    
72. [148220.026566] usb 1-2.1: Found UVC 1.00 device USB 2.0 Camera (0c45:6362)
    
73. [148220.054790] usbcore: registered new interface driver uvcvideo
    
74. [148220.079573] usbcore: registered new interface driver snd-usb-audio
    
75. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ ls -la /dev/video*
    
76. crw-rw----+ 1 root video 81, 19 Apr 28 11:25 /dev/video0
    
77. crw-rw----+ 1 root video 81, 20 Apr 28 11:25 /dev/video1
    
78. crw-rw----+ 1 root video 81, 18 Apr 26 18:15 /dev/video19
    
79. crw-rw----+ 1 root video 81,  0 Apr 26 18:15 /dev/video20
    
80. crw-rw----+ 1 root video 81,  1 Apr 26 18:15 /dev/video21
    
81. crw-rw----+ 1 root video 81,  2 Apr 26 18:15 /dev/video22
    
82. crw-rw----+ 1 root video 81,  3 Apr 26 18:15 /dev/video23
    
83. crw-rw----+ 1 root video 81,  4 Apr 26 18:15 /dev/video24
    
84. crw-rw----+ 1 root video 81,  5 Apr 26 18:15 /dev/video25
    
85. crw-rw----+ 1 root video 81,  6 Apr 26 18:15 /dev/video26
    
86. crw-rw----+ 1 root video 81,  7 Apr 26 18:15 /dev/video27
    
87. crw-rw----+ 1 root video 81,  8 Apr 26 18:15 /dev/video28
    
88. crw-rw----+ 1 root video 81,  9 Apr 26 18:15 /dev/video29
    
89. crw-rw----+ 1 root video 81, 10 Apr 26 18:15 /dev/video30
    
90. crw-rw----+ 1 root video 81, 11 Apr 26 18:15 /dev/video31
    
91. crw-rw----+ 1 root video 81, 12 Apr 26 18:15 /dev/video32
    
92. crw-rw----+ 1 root video 81, 13 Apr 26 18:15 /dev/video33
    
93. crw-rw----+ 1 root video 81, 14 Apr 26 18:15 /dev/video34
    
94. crw-rw----+ 1 root video 81, 15 Apr 26 18:15 /dev/video35
    
95. crw-rw----+ 1 root video 81, 16 Apr 26 18:15 /dev/video36
    
96. crw-rw----+ 1 root video 81, 17 Apr 26 18:15 /dev/video37
    
97. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ libcamera-hello --list-cameras
    
98. No cameras available!
    
99. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ v4l2-ctl --list-devices
    
100. pispbe (platform:1000880000.pisp_be):
     
101.         /dev/video20
     
102.         /dev/video21
     
103.         /dev/video22
     
104.         /dev/video23
     
105.         /dev/video24
     
106.         /dev/video25
     
107.         /dev/video26
     
108.         /dev/video27
     
109.         /dev/video28
     
110.         /dev/video29
     
111.         /dev/video30
     
112.         /dev/video31
     
113.         /dev/video32
     
114.         /dev/video33
     
115.         /dev/video34
     
116.         /dev/video35
     
117.         /dev/video36
     
118.         /dev/video37
     
119.         /dev/media0
     
120.         /dev/media1
     
121. 
     
122. rpivid (platform:rpivid):
     
123.         /dev/video19
     
124.         /dev/media2
     
125. 
     
126. USB 2.0 Camera: USB Camera (usb-xhci-hcd.0-2.1):
     
127.         /dev/video0
     
128.         /dev/video1
     
129.         /dev/media3
     
130. 
     
131. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ media-ctl -d /dev/media3 -p
     
132. Media controller API version 6.6.20
     
133. 
     
134. Media device information
     
135. ------------------------
     
136. driver          uvcvideo
     
137. model           USB 2.0 Camera: USB Camera
     
138. serial
     
139. bus info        usb-xhci-hcd.0-2.1
     
140. hw revision     0x0
     
141. driver version  6.6.20
     
142. 
     
143. Device topology
     
144. - entity 1: USB 2.0 Camera: USB Camera (1 pad, 1 link)
     
145.             type Node subtype V4L flags 1
     
146.             device node name /dev/video0
     
147.         pad0: Sink
     
148.                 <- "Extension 3":1 [ENABLED,IMMUTABLE]
     
149. 
     
150. - entity 4: USB 2.0 Camera: USB Camera (0 pad, 0 link)
     
151.             type Node subtype V4L flags 0
     
152.             device node name /dev/video1
     
153. 
     
154. - entity 8: Extension 3 (2 pads, 2 links)
     
155.             type V4L2 subdev subtype Unknown flags 0
     
156.         pad0: Sink
     
157.                 <- "Processing 2":1 [ENABLED,IMMUTABLE]
     
158.         pad1: Source
     
159.                 -> "USB 2.0 Camera: USB Camera":0 [ENABLED,IMMUTABLE]
     
160. 
     
161. - entity 11: Processing 2 (2 pads, 2 links)
     
162.              type V4L2 subdev subtype Unknown flags 0
     
163.         pad0: Sink
     
164.                 <- "Camera 1":0 [ENABLED,IMMUTABLE]
     
165.         pad1: Source
     
166.                 -> "Extension 3":0 [ENABLED,IMMUTABLE]
     
167. 
     
168. - entity 14: Camera 1 (1 pad, 1 link)
     
169.              type V4L2 subdev subtype Sensor flags 0
     
170.         pad0: Source
     
171.                 -> "Processing 2":0 [ENABLED,IMMUTABLE]
     
172. 
     
173. zhaomeijing@raspberrypi5:~/workspace/09_uvc$

复制代码

//------USB摄像头状态

2.        USB摄像头测试

将USB摄像头插入开发板，将自动安装 uvc 添加设备节点。  

USB 2.0 Camera: USB Camera (usb-xhci-hcd.0-2.1):

/dev/video0/dev/video1/dev/media3

USB摄像头测试内容包括设备节点检测、视频、拍照，测试指令如下：

1. //------USB-Camera（某宝几十块买的USB摄像头，支持1080P）
   
2. a.确认USB摄像头的设备节点，此处为/dev/video0。
   
3. v4l2-ctl --list-devices
   
4. b.查看摄像头支持的格式与分辨率
   
5. v4l2-ctl --list-formats-ext -d /dev/video0
   
6. c.摄像头预览（视频640 x 480）
   
7. gst-launch-1.0 v4l2src device=/dev/video0 ! \
   
8. "video/x-raw, width=640, Height=480, framerate=(fraction)30/1" ! \
   
9. queue ! glimagesink
   
10. d.摄像头预览（视频1920 x 1080）
    
11. gst-launch-1.0 v4l2src device=/dev/video0 ! \
    
12. video/x-raw,format=YUY2,width=1920,height=1080 ! \
    
13. queue ! glimagesink
    
14. e.摄像头预览（拍照）
    
15. gst-launch-1.0 v4l2src device=/dev/video0 num-buffers=3 ! \
    
16. jpegenc ! filesink location=~/pic_of_video0.jpg

复制代码

//------测试场景

//------USB摄像头测试正常（此处展示的是视频捕捉的测试场景；需要注意，分辨率越高，需要的响应时间越长）

3.        边沿检测

USB摄像头的视频流主要通过python调用OpenCV库完成对每一帧图像的边沿检测，参数调节通过滑动条实时调节以获取更好结果，按下ESC键退出测试程序（不得不说python是真的强大）：

1. //------基于OPENCV的摄像头视频边沿检测
   
2. //---指令（python2或者python3均可以）：
   
3. python3 ./edge.py 0
   
4. 或者
   
5. python ./edge.py 0
   
6. //---指令中的参数”0“说明：
   
7. 使用USB-Camera，它位于/dev/video0
   
8. //------依赖（根据错误提示安装对应的依赖库及以来组件）:
   
9. sudo apt install opencv*
   
10. sudo apt install libopencv*
    
11. sudo apt install gststream*
    
12. sudo apt install python3-opencv*
    
13. //---Log信息：
    
14. zhaomeijing@raspberrypi5:~/workspace/09_uvc$ python3 ./edge.py 0
    
15. 
    
16. Video capture sample.
    
17. 
    
18. Sample shows how VideoCapture class can be used to acquire video
    
19. frames from a camera of a movie file. Also the sample provides
    
20. an example of procedural video generation by an object, mimicking
    
21. the VideoCapture interface (see Chess class).
    
22. 
    
23. 'create_capture' is a convenience function for capture creation,
    
24. falling back to procedural video in case of error.
    
25. 
    
26. Usage:
    
27.     video.py [--shotdir <shot path>] [source0] [source1] ...'
    
28. 
    
29.     sourceN is an
    
30.      - integer number for camera capture
    
31.      - name of video file
    
32.      - synth:<params> for procedural video
    
33. 
    
34. Synth examples:
    
35.     synth:bg=lena.jpg:noise=0.1
    
36.     synth:class=chess:bg=lena.jpg:noise=0.1:size=640x480
    
37. 
    
38. Keys:
    
39.     ESC    - exit
    
40.     SPACE  - save current frame to <shot path> directory
    
41. 
    
42. 
    
43. [ WARN:0@1.380] global ./modules/videoio/src/cap_gstreamer.cpp (1405) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
    
44. Done
    
45. zhaomeijing@raspberrypi5:~/workspace/09_uvc$

复制代码

3.1        代码解析

代码功能说明：

1. //------代码功能说明：
   
2. a.导入必要的库：cv2、numpy、video 和 sys。
   
3. b.创建两个滑动条：cv.namedWindow('edge') 用来显示边缘检测的结果窗口，cv.createTrackbar('thrs1', 'edge', 2000, 5000, nothing) 和 cv.createTrackbar('thrs2', 'edge', 4000, 5000, nothing) 用来设置边缘检测的参数。
   
4. c.创建一个视频捕获对象：cap = video.create_capture(fn)，其中 fn 是指定的视频文件名。
   
5. d.在一个无限循环中，从视频捕获对象中读取帧数据：while True:。
   
6. e.将读取的帧数据转换为灰度图像：gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)。
   
7. f.对灰度图像进行边缘检测：edge = cv.Canny(gray, thrs1, thrs2, apertureSize=5)。
   
8. g.将原始图像和边缘检测结果合并：vis = img.copy()，然后将结果图像中非零像素的部分设为绿色：vis = np.uint8(vis/2.)，最后将结果图像显示出来：cv.imshow('edge', vis)。
   
9. h.监听窗口的关闭事件：ch = cv.waitKey(5)，如果按下了ESC键，则退出循环：if ch == 27:。
   
10. i.最后打印程序的文档字符串：print(__doc__)，并关闭所有窗口：cv.destroyAllWindows()。

复制代码

//------完整代码

1. #!/usr/bin/env python
   
2. 
   
3. from __future__ import print_function
   
4. 
   
5. import cv2 as cv
   
6. import numpy as np
   
7. 
   
8. # relative module
   
9. import video
   
10. 
    
11. # built-in module
    
12. import sys
    
13. 
    
14. def main():
    
15.     try:
    
16.         fn = sys.argv[1]
    
17.     except:
    
18.         fn = 0
    
19. 
    
20.     def nothing(*arg):
    
21.         pass
    
22. 
    
23.     cv.namedWindow('edge')
    
24.     cv.createTrackbar('thrs1', 'edge', 2000, 5000, nothing)
    
25.     cv.createTrackbar('thrs2', 'edge', 4000, 5000, nothing)
    
26. 
    
27.     cap = video.create_capture(fn)
    
28.     while True:
    
29.         _flag, img = cap.read()
    
30.         gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
    
31.         thrs1 = cv.getTrackbarPos('thrs1', 'edge')
    
32.         thrs2 = cv.getTrackbarPos('thrs2', 'edge')
    
33.         edge = cv.Canny(gray, thrs1, thrs2, apertureSize=5)
    
34.         vis = img.copy()
    
35.         vis = np.uint8(vis/2.)
    
36.         vis[edge != 0] = (0, 255, 0)
    
37.         cv.imshow('edge', vis)
    
38.         ch = cv.waitKey(5)
    
39.         if ch == 27:
    
40.             break
    
41. 
    
42.     print('Done')
    
43. 
    
44. 
    
45. if __name__ == '__main__':
    
46.     print(__doc__)
    
47.     main()
    
48.     cv.destroyAllWindows()

复制代码

3.2        测试效果

基于USB-Camera的边沿检测测试结果如图所示：

//------end