@veye_xumm sorry for necro-posting, but this method stopped working. on fresh system I have Failed to set I2C address after running make.sh under mv_tools_rpi/sources on rpi5. do you know if something has changed there?

@lozinski Sorry, we currently do not have products that meet your requirements. The VEYE-LVDS-327 has been discontinued for a long time.

@mparem
https://github.com/veyeimaging/rk35xx_orangepi/releases/tag/v1.2_mvcam_tablet_cm5
Please try this image file on cm5-tablet.

@barbanevosa
The issue occurs because the camera sensor outputs a resolution of 3088 × 2064, but the receiving side (V4L2 buffers) aligns each line to a stride of 3136 bytes for memory alignment. When OpenCV is used directly with cv2.CAP_V4L2, it applies the width setting (3088) both to the sensor and to the buffer interpretation. Since the actual buffer has padding (3136 vs. 3088), OpenCV misinterprets the data layout, which results in visible image artifacts (stripes).

We recommend using a GStreamer pipeline to handle the stride and alignment internally, so that OpenCV receives a clean image with the correct resolution.

@veye_xumm 好的 一定

@jhaijun 不好意思我们这个论坛是只针对我们公司自己产品的。 sony 机芯的话确实没有搞过。

@goooofy 好的。

Your current image is Jetpack 5.1.2 with L4T version r35.4.1. The mvcam driver included in it is relatively old, so the mv_probe.sh script cannot work properly. — For more details, you can refer to the comments at the top of the mv_probe.sh script.

081b1526-6747-417e-af94-d33ffbbc08cb-image.png

In addition, please try running the following script and see if it shows continuous messages indicating that data is being received. If it still fails to receive data, please run sudo dmesg to check whether there are any error messages.

export WIDTH=2560 export HEIGHT=1544 export FPS=50 v4l2-ctl --set-ctrl roi_x=0 v4l2-ctl --set-ctrl roi_y=0 v4l2-ctl --set-fmt-video=width=$WIDTH,height=$HEIGHT,pixelformat=GREY v4l2-ctl --set-ctrl frame_rate=$FPS v4l2-ctl --set-fmt-video=width=$WIDTH,height=$HEIGHT,pixelformat=GREY --stream-mmap --stream-count=-1 --stream-to=/dev/null

@golden0514 你好。
我们相机有一个底层逻辑，fps参数会优先于曝光时间设置。
如果你想要1秒的曝光时间，需要先将帧率设置为1帧/秒，然后再设置曝光时间即可。

@veye_xumm 这样啊，好的麻烦您了，我再跟公司里沟通一下

@nian 好的

@nian 不客气

@linoooooo
你好，fps参数是可以修改的。但是需要先停止取图。曝光时间是被限制到1/fps的。 你可以先设置合适的fps，然后在范围内进行曝光时间的修改。

在这个脚本中，带星号的命令，都是取图中写保护的。
你也可以参考寄存器清单中的part1.3来理解这个设置规则。

@fumyang 请问您用的是哪个相机模组？

@veye_xumm Thanks! I'll read and reinstall the SO.
Hopefully, I'll be able to show if the problem is resolved in the next few days.

@dragos You are welcome.

@linoooooo 你好，我们的相机内部isp流程是没有整帧缓存的，相机本身引入的delay是毫秒之内。
我觉得你这个延时主要是v4l2取图缓存和显示缓存的缘故。不过说实话SDL库我们不太懂，怎样优化需要您自己研究一下。
至于相机采集后，您这边程序直接导入算法的话，延迟肯定要比这种测量方法小的多。

@otca
The camera is now successfully probed, which confirms that both the I2C and the enable pin are functioning correctly.
The current issue indicates that the I2C control channel is working, but the MIPI CSI data channel is not. Please verify whether the MIPI CSI interface pin assignment on this carrier board matches that of the official Orin NX development board.
The Device Tree Source (DTS) should be adjusted accordingly to match the carrier board.

@stiwana Sorry, we do have this in mind, but no actual plans or projected time of completion at this time.

@阿坎德 你好，有进一步的结果吗？