Forum - Recent Threads

I am getting only 1 channel out of 2, for each serial pin in Beaglebone AI

mqya95 — Mon, 21 Mar 2022 09:53:32 GMT

Hello I want to record 8 channels from TIDA-01454 CMB into a Beaglebone AI. As the CMB is built with two PCM1864 ADCs and it is also a Beagle board, I followed this guide(https://www.ti.com/lit/an/sprac97/sprac97.pdf) with some changes(channels_max=16).

I have managed to record audio from 4 of the 8 microphones that the CMB has(I just tap the mic to check if that one is working). However I want to record the 8 channels. Currently the working microphones are MIC1, MIC4, MIC5 and MIC8, although I would say there is much noise.The CMB has 4 data output pins, I suppose each one transmits 2 channels, so the problem is that each pin is getting mono instead of stereo. This might be because a problem in DTS, therefore this is my dts file:

[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:ced9c8e6-d313-4ebd-8165-95018374148d:type=text&text=pcm5102a%3A%20pcm5102a%20%7B%20%0D%0A%0D%0A%20%20%20%20%20%23sound-dai-cells%20%3D%20%3C0%3E%3B%20%0D%0A%0D%0A%20%20%20%20%20compatible%20%3D%20%22ti%2Cpcm5102a%22%3B%20%0D%0A%0D%0A%20%20%20%20%20status%20%3D%20%22okay%22%3B%20%0D%0A%0D%0A%20%20%7D%3B%20%0D%0A%0D%0A%20%20%20%0D%0A%0D%0A%20%20%20%0D%0A%0D%0A%20%20sound%20%7Bcompatible%20%3D%20%22simple-audio-card%22%3B%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Cformat%20%3D%20%22i2s%22%3B%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Cname%20%3D%20%22PCM5102a%22%3B%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Cbitclock-master%20%3D%20%3C%26sound1_master%3E%3B%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Cframe-master%20%3D%20%3C%26sound1_master%3E%3B%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Cbitclock-inversion%3B%20%0D%0A%0D%0A%20%20%20%20%20%0D%0A%0D%0A%20%20%20%20%20%0D%0A%0D%0A%20%20%20%20simple-audio-card%2Ccpu%20%7B%20%0D%0A%0D%0A%20%20%20%20%20%20%20%20%20%20%20%20sound-dai%20%3D%20%3C%26mcasp1%3E%3B%20%0D%0A%0D%0A%20%20%20%20%7D%3B%20%0D%0A%0D%0A%20%20%20%20sound1_master%3A%20simple-audio-card%2Ccodec%20%7B%20%0D%0A%0D%0A%20%20%20%20%20%20%20%20%23sound-dai-cells%20%3D%20%3C0%3E%3B%20%0D%0A%0D%0A%20%20%20%20%20%20%20%20sound-dai%20%3D%20%3C%26pcm5102a%3E%3B%20%0D%0A%0D%0A%20%20%20%20%20%20%20%20%20%20%2F%2Fclocks%20%3D%20%3C%26mcasp1_fck%3E%3B%20%0D%0A%0D%0A%20%20%20%20%20%20%20%20%20%20%2F%2Fclock-names%20%3D%20%22mclk%22%3B%20%0D%0A%0D%0A%20%20%20%20%20%20%7D%3B%20%0D%0A%0D%0A%20%20%7D%3B%20%0D%0A%0D%0A%26mcasp1%20%7B%20%0D%0A%0D%0A%20%20%20%20%23sound-dai-cells%20%3D%20%3C0%3E%3B%20%0D%0A%0D%0A%20%20%20%20pinctrl-names%20%3D%20%22default%22%3B%20%0D%0A%0D%0A%20%20%20%20pinctrl-0%20%3D%20%3C%26mcasp1_pins%3E%3B%20%0D%0A%0D%0A%20%20%20%20status%20%3D%20%22okay%22%3B%20%0D%0A%0D%0A%20%20%20%20op-mode%20%3D%20%3C0%3E%3B%20%2F%2A%20MCASP_IIS_MODE%20%2A%2F%20%0D%0A%0D%0A%20%20%20%20tdm-slots%20%3D%20%3C2%3E%3B%20%0D%0A%0D%0A%20%20%20%20num-serializer%20%3D%20%3C4%3E%3B%20%0D%0A%0D%0A%20%20%20%20serial-dir%20%3D%20%3C%20%2F%2A%201%20TX%202%20RX%200%20unused%20%2A%2F%20%0D%0A%0D%0A%20%20%20%20%20%20%20%202%202%200%200%200%200%200%200%200%200%202%202%0D%0A%0D%0A%20%20%20%20%20%20%3E%3B%20%0D%0A%0D%0A%20%20%20%20rx-num-evt%20%3D%20%3C4%3E%3B%20%0D%0A%0D%0A%20%20%20%20tx-num-evt%20%3D%20%3C4%3E%3B%20%0D%0A%0D%0A%7D%3B]

The serial-dir is that way because I use mcasp1_axr0, mcasp1_axr1, mcasp1_10 and mcasp1_axr11 ,those are the ones available in Beaglbone AI. I connected the pins DATA1-MiniDSP, DATA2-MiniDSP, DATA3-MiniDSP and DATA4-MiniDSP in J5(of the CMB) to these pins in my Beaglebone:


[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:fb1b3272-85f8-426a-9639-afe855b3f24d:type=text&text=P9.18b%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20173%20fast%20rx%20%20%20%20%20%20%200%20mcasp%200%20d0%20%20%20%20%20%20%20mcasp%4048460000%20%28mcasp1_pins%29%0D%0AP9.17b%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20174%20fast%20rx%20%20%20%20%20%20%200%20mcasp%200%20d1%20%20%20%20%20%20%20mcasp%4048460000%20%28mcasp1_pins%29%0D%0AP9.30%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20183%20fast%20rx%20%20%20%20%20%20%200%20mcasp%200%20d10%20%20%20%20%20%20mcasp%4048460000%20%28mcasp1_pins%29%0D%0AP9.28%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20184%20fast%20rx%20%20%20%20%20%20%200%20mcasp%200%20d11%20%20%20%20%20%20mcasp%4048460000%20%28mcasp1_pins%29]

I also conected DATA1-MiniDSP, LRCLK and BCLK to an external I2S DAC to make a test and I could hear MIC1 and MIC2. So I can assure that outputs are coming in stereo. My problem is that when connect the outputs to Beaglebone and record, I just get 1 of those 2 channels for each DOUT. I checked all of them just in case with the external I2S DAC:

Data 1 is giving MIC1 and MIC2

Data 2 is giving MIC5 and MIC6

Data 3 is giving MIC3 and MIC4

Data 4 is giving MIC7 and MIC8

In case it helps, here is myDummy CODEC:

[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:e642bf31-726c-477b-a755-b6d51ba1f043:type=text&text=%23include%20%3Clinux%2Finit.h%3E%0D%0A%23include%20%3Clinux%2Fmodule.h%3E%0D%0A%23include%20%3Clinux%2Fplatform_device.h%3E%0D%0A%0D%0A%23include%20%3Csound%2Fsoc.h%3E%0D%0A%0D%0Astatic%20struct%20snd_soc_dai_driver%20pcm5102a_dai%20%3D%20%7B%0D%0A%09.name%20%3D%20%22pcm5102a-hifi%22%2C%0D%0A%09.playback%20%3D%20%7B%0D%0A%09%09.channels_min%20%3D%202%2C%0D%0A%09%09.channels_max%20%3D%202%2C%0D%0A%09%09.rates%20%3D%20SNDRV_PCM_RATE_8000_192000%2C%0D%0A%09%09.formats%20%3D%20SNDRV_PCM_FMTBIT_S16_LE%20%7C%0D%0A%09%09%09%20%20%20SNDRV_PCM_FMTBIT_S24_LE%20%7C%0D%0A%09%09%09%20%20%20SNDRV_PCM_FMTBIT_S32_LE%0D%0A%09%7D%2C%0D%0A%09.capture%20%3D%20%7B%20%0D%0A%0D%0A%09%09.stream_name%20%3D%20%22Capture%22%2C%20%0D%0A%0D%0A%09%09.channels_min%20%3D%201%2C%20%0D%0A%0D%0A%09%09.channels_max%20%3D%2016%2C%20%0D%0A%0D%0A%09%09.rates%20%3D%20SNDRV_PCM_RATE_8000_192000%2C%20%0D%0A%0D%0A%09%09.formats%20%3D%20SNDRV_PCM_FMTBIT_S16_LE%20%7C%20%0D%0A%0D%0A%09%09%09SNDRV_PCM_FMTBIT_S24_LE%20%7C%20%0D%0A%0D%0A%09%09%09SNDRV_PCM_FMTBIT_S32_LE%20%0D%0A%0D%0A%09%7D%2C%20%0D%0A%7D%3B%0D%0A%0D%0Astatic%20struct%20snd_soc_codec_driver%20soc_codec_dev_pcm5102a%3B%0D%0A%0D%0Astatic%20int%20pcm5102a_probe%28struct%20platform_device%20%2Apdev%29%0D%0A%7B%0D%0A%09return%20snd_soc_register_codec%28%26pdev-%3Edev%2C%20%26soc_codec_dev_pcm5102a%2C%0D%0A%09%09%09%26pcm5102a_dai%2C%201%29%3B%0D%0A%7D%0D%0A%0D%0Astatic%20int%20pcm5102a_remove%28struct%20platform_device%20%2Apdev%29%0D%0A%7B%0D%0A%09snd_soc_unregister_codec%28%26pdev-%3Edev%29%3B%0D%0A%09return%200%3B%0D%0A%7D%0D%0A%0D%0Astatic%20const%20struct%20of_device_id%20pcm5102a_of_match%5B%5D%20%3D%20%7B%0D%0A%09%7B%20.compatible%20%3D%20%22ti%2Cpcm5102a%22%2C%20%7D%2C%0D%0A%09%7B%20%7D%0D%0A%7D%3B%0D%0AMODULE_DEVICE_TABLE%28of%2C%20pcm5102a_of_match%29%3B%0D%0A%0D%0Astatic%20struct%20platform_driver%20pcm5102a_codec_driver%20%3D%20%7B%0D%0A%09.probe%09%09%3D%20pcm5102a_probe%2C%0D%0A%09.remove%09%09%3D%20pcm5102a_remove%2C%0D%0A%09.driver%09%09%3D%20%7B%0D%0A%09%09.name%09%3D%20%22pcm5102a-codec%22%2C%0D%0A%09%09.of_match_table%20%3D%20pcm5102a_of_match%2C%0D%0A%09%7D%2C%0D%0A%7D%3B%0D%0A%0D%0Amodule_platform_driver%28pcm5102a_codec_driver%29%3B%0D%0A%0D%0AMODULE_DESCRIPTION%28%22ASoC%20PCM5102A%20codec%20driver%22%29%3B%0D%0AMODULE_AUTHOR%28%22Florian%20Meier%20%3Cflorian.meier%40koalo.de%3E%22%29%3B%0D%0AMODULE_LICENSE%28%22GPL%20v2%22%29%3B]

and also my CMB configuration:

[embed:dc8ab71f-3b98-42d9-b0f6-e21e02a0f8e2:90425798-5b0a-4e72-a494-b4d169c9c9de:type=text&text=uint8_t%20U1_PCM1864_CONFIG%5B%5D%5B2%5D%20%3D%20%7B%20%0D%0A%0D%0A%7B0x00%2C%200x00%7D%2C%20%2F%2F%20Change%20to%20Page%200%20%0D%0A%0D%0A%7B0x01%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH1_L%20to%2032dB%20%0D%0A%0D%0A%7B0x02%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH1_R%20to%2032dB%20%0D%0A%0D%0A%7B0x03%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH2_L%20to%2032dB%20%0D%0A%0D%0A%7B0x04%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH2_R%20to%2032dB%20%0D%0A%0D%0A%7B0x05%2C%200x86%7D%2C%20%2F%2F%20Enable%20SMOOTH%20PGA%20Change%3B%20Independent%20Link%20PGA%3B%20%0D%0A%0D%0A%7B0x06%2C%200x41%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINL1%5BSE%5D%20%0D%0A%0D%0A%7B0x07%2C%200x41%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINR1%5BSE%5D%20%0D%0A%0D%0A%7B0x08%2C%200x44%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINL3%5BSE%5D%20%0D%0A%0D%0A%7B0x09%2C%200x44%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINR3%5BSE%5D%20%0D%0A%0D%0A%7B0x0A%2C%200x00%7D%2C%20%2F%2F%20Secondary%20ADC%20Input%3A%20No%20Selection%20%0D%0A%0D%0A%7B0x0B%2C%200x44%7D%2C%20%2F%2F%20RX%20WLEN%3A%2024bit%3B%20TX%20WLEN%3A%2024%20bit%3B%20FMT%3A%20I2S%20format%20%0D%0A%0D%0A%7B0x10%2C%200x03%7D%2C%20%2F%2F%20GPIO0_FUNC%20-%20SCK%20Out%3B%20GPIO0_POL%20-%20Normal%20%0D%0A%0D%0A%7B0x11%2C%200x50%7D%2C%20%2F%2F%20GPIO3_FUNC%20-%20DOUT2%3B%20GPIO3_POL%20-%20Normal%20%0D%0A%0D%0A%7B0x12%2C%200x04%7D%2C%20%2F%2F%20GPIO0_DIR%20-%20GPIO0%20-%20Output%20%0D%0A%0D%0A%7B0x13%2C%200x40%7D%2C%20%2F%2F%20GPIO3_DIR%20%EF%BF%BD%20GPIO3%20-%20Output%20%0D%0A%0D%0A%7B0x20%2C%200x11%7D%20%20%2F%2F%20MST_MODE%3A%20Master%3B%20CLKDET_EN%3A%20Disable%20%0D%0A%0D%0A%7D%3B%20%0D%0A%0D%0A%20%20%0D%0A%0D%0Auint8_t%20U2_PCM1864_CONFIG%5B%5D%5B2%5D%20%3D%20%7B%20%0D%0A%0D%0A%7B0x00%2C%200x00%7D%2C%20%2F%2F%20Change%20to%20Page%200%20%0D%0A%0D%0A%7B0x01%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH1_L%20to%2032dB%20%0D%0A%0D%0A%7B0x02%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH1_R%20to%2032dB%20%0D%0A%0D%0A%7B0x03%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH2_L%20to%2032dB%20%0D%0A%0D%0A%7B0x04%2C%200x40%7D%2C%20%2F%2F%20PGA%20CH2_R%20to%2032dB%20%0D%0A%0D%0A%7B0x05%2C%200x86%7D%2C%20%2F%2F%20Enable%20SMOOTH%20PGA%20Change%3B%20Independent%20Link%20PGA%3B%20%0D%0A%0D%0A%7B0x06%2C%200x41%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINL1%5BSE%5D%20%0D%0A%0D%0A%7B0x07%2C%200x41%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINR1%5BSE%5D%20%0D%0A%0D%0A%7B0x08%2C%200x44%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINL3%5BSE%5D%20%0D%0A%0D%0A%7B0x09%2C%200x44%7D%2C%20%2F%2F%20Polarity%3A%20Normal%2C%20Channel%3A%20VINR3%5BSE%5D%20%0D%0A%0D%0A%7B0x0A%2C%200x00%7D%2C%20%2F%2F%20Secondary%20ADC%20Input%3A%20No%20Selection%20%0D%0A%0D%0A%7B0x0B%2C%200x44%7D%2C%20%2F%2F%20RX%20WLEN%3A%2024bit%3B%20TX%20WLEN%3A%2024%20bit%3B%20FMT%3A%20I2S%20format%20%0D%0A%0D%0A%7B0x10%2C%200x00%7D%2C%20%2F%2F%20GPIO0_FUNC%20%E2%80%93%20GPIO0%3B%20GPIO0_POL%20-%20Normal%20%0D%0A%0D%0A%7B0x11%2C%200x50%7D%2C%20%2F%2F%20GPIO3_FUNC%20-%20DOUT2%3B%20GPIO3_POL%20-%20Normal%20%0D%0A%0D%0A%7B0x12%2C%200x00%7D%2C%20%2F%2F%20GPIO0_DIR%20-%20GPIO0%20-%20Input%20%0D%0A%0D%0A%7B0x13%2C%200x40%7D%2C%20%2F%2F%20GPIO3_DIR%20%EF%BF%BD%20GPIO3%20-%20Output%20%0D%0A%0D%0A%7B0x20%2C%200x01%7D%20%20%2F%2F%20MST_MODE%3A%20Slave%3B%20CLKDET_EN%3A%20Enable%20%0D%0A%0D%0A%7D%3B]

So what is wrong? How can I get all the mics?

AI training online

clem57 — Thu, 05 Dec 2019 20:31:19 GMT

For anyone is interested, there is now a free AI course by IBM using Coursea platform. Since it is delivered online, you can partake as you wish around the world. Check it out at this topic thread .

Clem

BB AI and OpenCV 3.3+

vlasov01 — Fri, 25 Oct 2019 02:09:45 GMT

I know BB AI comes with OpenCV 3.2.

I'd like to use BB AI with OpenCV Deep Neural Networks (dnn module) on my project. It comes with OpenCV 3.3 or later.

I'm wondering if anyone was able to get OpenCV 3.3 or later for BB AI?

If yes, are you still benefiting from BB AI hardware accelerators?

RE: BB AI and OpenCV 3.3+

vlasov01 — Mon, 04 Nov 2019 01:52:26 GMT

Hi Adrian,

I'm using virtual python environment. And it was setup only for user debian. So when I run the command you suggested it is running it using a different version of OpenCV:

sudo python -c "import cv2; print(cv2.__version__)"
3.2.0

And I was looking for OpenCV 3.3+ as it has support for dnn.

I've tried another approach:

cd /home/debian/.virtualenvs/cv/bin/
sudo setcap cap_sys_rawio+ep python3

But it gave me another error:

******************************************************************
* FATAL ERROR:                                                   *
* This OpenCV build doesn't support current CPU/HW configuration *
*                                                                *
* Use OPENCV_DUMP_CONFIG=1 environment variable for details      *
******************************************************************

Required baseline features:
    ID=100 (NEON) - NOT AVAILABLE
terminate called after throwing an instance of 'cv::Exception'
  what():  OpenCV(4.1.2) /home/debian/opencv/modules/core/src/system.cpp:598: error: (-215:Assertion failed) Missing support for required CPU baseline features. Check OpenCV build configuration and required CPU/HW setup. in function 'initialize'

Aborted

And if I set OPENCV_DUMP_CONFIG=1 then I'm getting the followimng information about my OpenCV configuration:

OpenCV build configuration is:

General configuration for OpenCV 4.1.2 =====================================
  Version control:               unknown

  Extra modules:
    Location (extra):            /home/debian/opencv_contrib/modules
    Version control (extra):     unknown

  Platform:
    Timestamp:                   2019-10-27T02:57:31Z
    Host:                        Linux 4.14.108-ti-r122 armv7l
    CMake:                       3.7.2
    CMake generator:             Unix Makefiles
    CMake build tool:            /usr/bin/make
    Configuration:               RELEASE

  CPU/HW features:
    Baseline:                    NEON
      requested:                 DETECT
      required:                  NEON
      disabled:                  VFPV3

  C/C++:
    Built as dynamic libs?:      YES
    C++ Compiler:                /usr/bin/c++  (ver 6.3.0)
    C++ flags (Release):         -fsigned-char -W -Wall -Werror=return-type -Werror=non-virtual-dtor -Werror=address -Werror=sequence-point -Wformat -Werror=format-security -Wmissing-declarations -Wundef -Winit-self -Wpointer-arith -Wshadow -Wsign-promo -Wuninitialized -Winit-self -Wno-psabi -Wsuggest-override -Wno-delete-non-virtual-dtor -Wno-comment -fdiagnostics-show-option -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections  -mfpu=neon -fvisibility=hidden -fvisibility-inlines-hidden -O3 -DNDEBUG  -DNDEBUG
    C++ flags (Debug):           -fsigned-char -W -Wall -Werror=return-type -Werror=non-virtual-dtor -Werror=address -Werror=sequence-point -Wformat -Werror=format-security -Wmissing-declarations -Wundef -Winit-self -Wpointer-arith -Wshadow -Wsign-promo -Wuninitialized -Winit-self -Wno-psabi -Wsuggest-override -Wno-delete-non-virtual-dtor -Wno-comment -fdiagnostics-show-option -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections  -mfpu=neon -fvisibility=hidden -fvisibility-inlines-hidden -g  -O0 -DDEBUG -D_DEBUG
    C Compiler:                  /usr/bin/cc
    C flags (Release):           -fsigned-char -W -Wall -Werror=return-type -Werror=non-virtual-dtor -Werror=address -Werror=sequence-point -Wformat -Werror=format-security -Wmissing-declarations -Wmissing-prototypes -Wstrict-prototypes -Wundef -Winit-self -Wpointer-arith -Wshadow -Wuninitialized -Winit-self -Wno-psabi -Wno-comment -fdiagnostics-show-option -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections  -mfpu=neon -fvisibility=hidden -O3 -DNDEBUG  -DNDEBUG
    C flags (Debug):             -fsigned-char -W -Wall -Werror=return-type -Werror=non-virtual-dtor -Werror=address -Werror=sequence-point -Wformat -Werror=format-security -Wmissing-declarations -Wmissing-prototypes -Wstrict-prototypes -Wundef -Winit-self -Wpointer-arith -Wshadow -Wuninitialized -Winit-self -Wno-psabi -Wno-comment -fdiagnostics-show-option -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections  -mfpu=neon -fvisibility=hidden -g  -O0 -DDEBUG -D_DEBUG
    Linker flags (Release):      -Wl,--gc-sections
    Linker flags (Debug):        -Wl,--gc-sections
    ccache:                      NO
    Precompiled headers:         NO
    Extra dependencies:          dl m pthread rt
    3rdparty dependencies:

  OpenCV modules:
    To be built:                 aruco bgsegm bioinspired calib3d ccalib core datasets dnn dnn_objdetect dnn_superres dpm face features2d flann freetype fuzzy gapi hfs highgui img_hash imgcodecs imgproc line_descriptor ml objdetect optflow phase_unwrapping photo plot python3 quality reg rgbd saliency shape stereo stitching structured_light superres surface_matching text tracking ts video videoio videostab xfeatures2d ximgproc xobjdetect xphoto
    Disabled:                    world
    Disabled by dependency:      -
    Unavailable:                 cnn_3dobj cudaarithm cudabgsegm cudacodec cudafeatures2d cudafilters cudaimgproc cudalegacy cudaobjdetect cudaoptflow cudastereo cudawarping cudev cvv hdf java js matlab ovis python2 sfm viz
    Applications:                perf_tests apps
    Documentation:               NO
    Non-free algorithms:         YES

  GUI:
    GTK+:                        YES (ver 3.22.11)
      GThread :                  YES (ver 2.50.3)
      GtkGlExt:                  NO
    VTK support:                 NO

  Media I/O:
    ZLib:                        /usr/lib/arm-linux-gnueabihf/libz.so (ver 1.2.8)
    JPEG:                        /usr/lib/arm-linux-gnueabihf/libjpeg.so (ver 62)
    WEBP:                        build (ver encoder: 0x020e)
    PNG:                         /usr/lib/arm-linux-gnueabihf/libpng.so (ver 1.6.28)
    TIFF:                        /usr/lib/arm-linux-gnueabihf/libtiff.so (ver 42 / 4.0.8)
    JPEG 2000:                   build (ver 1.900.1)
    OpenEXR:                     /usr/lib/arm-linux-gnueabihf/libImath.so /usr/lib/arm-linux-gnueabihf/libIlmImf.so /usr/lib/arm-linux-gnueabihf/libIex.so /usr/lib/arm-linux-gnueabihf/libHalf.so /usr/lib/arm-linux-gnueabihf/libIlmThread.so (ver 2.2.0)
    HDR:                         YES
    SUNRASTER:                   YES
    PXM:                         YES
    PFM:                         YES

  Video I/O:
    DC1394:                      YES (2.2.5)
    FFMPEG:                      YES
      avcodec:                   YES (57.64.101)
      avformat:                  YES (57.56.101)
      avutil:                    YES (55.34.101)
      swscale:                   YES (4.2.100)
      avresample:                YES (3.1.0)
    GStreamer:                   NO
    v4l/v4l2:                    YES (linux/videodev2.h)

  Parallel framework:            pthreads

  Trace:                         YES (with Intel ITT)

  Other third-party libraries:
    Lapack:                      NO
    Eigen:                       NO
    Custom HAL:                  YES (carotene (ver 0.0.1))
    Protobuf:                    build (3.5.1)

  OpenCL:                        YES (no extra features)
    Include path:                /home/debian/opencv/3rdparty/include/opencl/1.2
    Link libraries:              Dynamic load

  Python 3:
    Interpreter:                 /home/debian/.virtualenvs/cv/bin/python3 (ver 3.5.3)
    Libraries:                   /usr/lib/arm-linux-gnueabihf/libpython3.5m.so (ver 3.5.3)
    numpy:                       /home/debian/.virtualenvs/cv/lib/python3.5/site-packages/numpy/core/include (ver 1.17.3)
    install path:                /home/debian/.virtualenvs/cv/lib/python3.5/site-packages/cv2/python-3.5

  Python (for build):            /usr/bin/python2.7

  Java:
    ant:                         NO
    JNI:                         NO
    Java wrappers:               NO
    Java tests:                  NO

  Install to:                    /usr/local
-----------------------------------------------------------------

RE: BB AI and OpenCV 3.3+

clem57 — Sun, 03 Nov 2019 19:20:58 GMT

I was really thing

and next line your command under root of course.

But sudo also...

RE: BB AI and OpenCV 3.3+

ajcc — Sun, 03 Nov 2019 17:02:23 GMT

Arguments given to su will be interpreted as usernames, you're thinking of sudo

sudo python -c "import cv2; print(cv2.__version__)"

RE: BB AI and OpenCV 3.3+

clem57 — Sun, 03 Nov 2019 15:15:20 GMT

Use su before python command.

RE: BB AI and OpenCV 3.3+

vlasov01 — Sun, 03 Nov 2019 11:37:05 GMT

I was able to build OpenCV 4.1.2 on BB AI.

(cv) debian@beaglebone:~$ python -c "import cv2; print(cv2.__version__)"
4.1.2

I'm not sure yet if it is using accelerators as I'm getting "TIOCL FATAL: Failed to open file /dev/mem" error when I'm calling dnn module from python.

RE: Installing OpenCV is Like...

yosoufe — Sun, 27 Oct 2019 05:38:51 GMT

What about cross-compile with docker container? i haven't done it but it may work

I found a page for cross-compiling for raspberry pi in docker container here: https://desertbot.io/blog/how-to-cross-compile-for-raspberry-pi

RE: Installing OpenCV is Like...

vlasov01 — Sat, 26 Oct 2019 16:30:34 GMT

Thank you for the reply!

I'm building now OpenCV 4.1.2 on BB AI. So I'll get my answers soon.

Installing OpenCV is Like...

14rhb — Wed, 09 Oct 2019 09:00:13 GMT

After several days into attempting to install OpenCV on my BB-AI it feels like I'm at about 0:15s into this video !

[View:/resized-image/__size/1250x750/__key/communityserver-discussions-components-files/251/2046.contentimage_5F00_121370.jpg:620:372]

https://www.youtube.com/watch?v=wmFzGXy5uYc

Maybe some pointers for others are in my iBox posts:

iBox: #2 Setting up the BB-AI

iBox: #3 Is it a cat, a plane, a moth?

I also tried cross compiling on my faster PC with the aim of porting the code across afterwards, but that too failed with the libpython2-dev:armhf library not being found. After much adjustment of my machine on the Linux command line I'm also no nearer on that approach.

But it did make me think, hence this question rather than a blog post:

If I were to compile the OpenCV on my other machine and port it across to the BB-AI then has someone else done this? e.g. why am I sitting waiting days for the BB-AI to run CMAKE if it could be done externally and highly likely someone already has.

I've learnt many Linux commands along the way and have a rough understanding of what cross-compiling is.....so progress todate has been worthwhile still [View:/resized-image/__size/16x16/__key/communityserver-discussions-components-files/251/4265.contentimage_5F00_1.png:16:16]

RE: Installing OpenCV is Like...

skywalker1211 — Sat, 26 Oct 2019 15:05:49 GMT

i don't have a BB AI. i install opencv4 into raspi2.

RE: Installing OpenCV is Like...

vlasov01 — Fri, 25 Oct 2019 23:11:30 GMT

I'm wondering if you noticed any performance differences between supplied OpenCV 3.2 and v4 for image processing tasks? Is v4 still leveraging DSP and EVE cores of AM5729?

RE: Installing OpenCV is Like...

skywalker1211 — Fri, 25 Oct 2019 15:43:12 GMT

14rhb

i changed from 100 -> 1024, and revert after installation.

RE: Installing OpenCV is Like...

14rhb — Fri, 25 Oct 2019 15:33:42 GMT

Thank you for this useful information skywalker1211 - do you have any recommendations for what you changed the swap space to?

RE: Installing OpenCV is Like...

skywalker1211 — Fri, 25 Oct 2019 13:41:21 GMT

there is no error message. i got the same problem when i compile opencv4 in raspi2. increase swap space solve the problem.

https://linuxize.com/post/how-to-install-opencv-on-raspberry-pi/ ---> "Start by increasing the swap space to avoid compilation hangups due to memory problems:"

RE: Beaglebone Ai Heating Issues

phoenixcomm — Thu, 24 Oct 2019 23:12:26 GMT

tariq.ahmad yes the fan and heat sink are issues. I will be putting this in a larger box that I will print. with a real CPU fan on one side and that should help with the cooling.

RE: Beaglebone Ai Heating Issues

tariq.ahmad — Wed, 23 Oct 2019 22:35:38 GMT

Thanks phoenixcomm , haven't heard a whole lot Glad to you see you are up and running again. Will be interesting to see how others use the fan or what solutions people come up with. Interesting to discussion on fans here as well:

First of All, Tame the roaming beast, BBB AI.

But looks like this has become the definitive post on bb ai heating issues...

RE: Beaglebone Ai Heating Issues

phoenixcomm — Wed, 23 Oct 2019 22:26:26 GMT

tariq.ahmad tarriq yea I had problems with the fan.. 5v @ 500ma and it you would see 1.45 volts...ouch. So I tossed the fan on my HP power supply set to .5amps and took it up from Ov through 5v about 8volts it started turning. and then took it up to 12volts. Now works fine on my 5v@500ma wall wart. LOL

My plan is to print a case for it and put a fan in the case as well. BTW if you put a spacer (1-2mm) between the fan and the heat sink you will get more airflow. As there will be less resistance on the fan and will pull more heat out as well.

RE: Beaglebone Ai Heating Issues

14rhb — Wed, 23 Oct 2019 21:20:37 GMT

RE: Beaglebone Ai Heating Issues

tariq.ahmad — Wed, 23 Oct 2019 21:14:20 GMT

Hi ipv1 ,

I am having another beaglebone ai sent to you.

RE: Problems getting started or is it dead??

shabaz — Mon, 21 Oct 2019 16:03:20 GMT

Hi Chris,

I agree with the comments above too, there's quite a fair chance it could be the PSU. I too saw the same symptom (brief start-up, many LEDs flickering, and then nothing) when I used a new Samsung charger. Swapped over to an official Pi PSU from Farnell/Newark (I used a Pi 3 5.1V 2.5A supply, but the Pi 4 5.1V 3A supply would be the better choice) and then it worked.

RE: Beaglebone Ai Heating Issues

cstanton — Mon, 21 Oct 2019 14:59:32 GMT

Had you connected anything to the black pin headers before it became unresponsive?

RE: Beaglebone Ai Heating Issues

ipv1 — Mon, 21 Oct 2019 14:38:02 GMT

Tried that as well but not go. Does not seem to do anything anymore.

RE: Problems getting started or is it dead??

mp2100 — Mon, 21 Oct 2019 14:01:05 GMT

Do you know what version of USB 3 your motherboard supports? The original USB 3.0 only provides 900 mA, so it's only 4.5 watts, probably not enough to run the BB AI.

https://en.wikipedia.org/wiki/USB_3.0#Power_and_charging

From your short video, the LEDs briefly light up like a normal beaglebone, then stop. It could be it starts booting but then needs more amps and doesn't get them.

Some USB 3 ports can deliver a lot more power, but only if it meets the "power delivery spec".

https://en.wikipedia.org/wiki/USB_3.0#Power_and_charging