Skip to content

Latest commit

 

History

History
81 lines (60 loc) · 3.02 KB

camera.md

File metadata and controls

81 lines (60 loc) · 3.02 KB
Raw

Using camera

Applicable platforms: STM32MP257F-EV1, STM32MP135-DK.

Video from the camera can be tested in the demo configuration, with the yavta tool.

First you have to configure Video4Linux using the media-ctl command. The configuration depends on your camera:

IMX335 Sensor (MB1854 camera board)

# media-ctl -d "platform:dcmipp" --set-v4l2 "'imx335 0-001a':0[fmt:SRGGB10_1X10/2592x1940]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'stm32_csi2host.48020000.csi2hos':1[fmt:SRGGB10_1X10/2592x1940]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_main_isp':1[fmt:RGB888_1X24/2592x1940 field:none]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_main_postproc':0[compose:(0,0)/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_main_postproc':1[fmt:RGB565_2X8_LE/640x480]"
# export camera_dev=$(media-ctl -d "platform:dcmipp" -e "dcmipp_dump_capture")

GC2145 Sensor (MB1897 camera board)

# media-ctl -d "platform:dcmipp" --set-v4l2 "'gc2145 1-003c':0[fmt:RGB565_2X8_BE/640x480@1/30 field:none]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_input':0[fmt:RGB565_2X8_BE/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_input':1[fmt:RGB565_2X8_LE/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_dump_postproc':0[fmt:RGB565_2X8_LE/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_dump_postproc':1[crop:(80,104)/480x272]"
# export camera_dev=$(media-ctl -d "platform:dcmipp" -e "dcmipp_dump_capture")

OV5640 Sensor (MB1723 camera board)

You first need to apply some changes to the Device Tree of the STM32MP135-DK board, in order to change from the GC2145 sensor (default) to the OV5640. The changes consist in:

  1. Enabling the OV5640

  2. Disabling the GC2145

  3. Changing the description of the endpoints used for the camera pipeline to connect the CSI-to-parallel bridge to the OV5640 instead of the GC2145.

The changes are described in the patch at https://gist.github.com/tpetazzoni/63f6c29337111d64a2cf35ed8d42104b.

# media-ctl -d "platform:dcmipp" --set-v4l2 "'ov5640 1-003c':0[fmt:RGB565_2X8_LE/640x480@1/30 field:none]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_input':0[fmt:RGB565_2X8_LE/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_dump_postproc':0[fmt:RGB565_2X8_LE/640x480]"
# media-ctl -d "platform:dcmipp" --set-v4l2 "'dcmipp_dump_postproc':1[crop:(80,104)/480x272]"
# export camera_dev=$(media-ctl -d "platform:dcmipp" -e "dcmipp_dump_capture")

Capturing frames

Use for example the Yavta tool, which is integrated into the demo root filesystem:

# yavta -F $camera_dev --capture=10

This command captures 10 frame from the camera with the default format and resolution.

You can list the available formats with the following command:

# yavta -l --enum-formats --enum-inputs $camera_dev

$ mv frame-000000.bin frame-000000.rgb
$ ffmpeg -s 640x480 -pix_fmt rgb565le -i frame-000000.rgb -f image2 -pix_fmt rgb24 frame0.png

Run these commands on your host computer to convert the rgb frame to a png image.