3

USB is a standard and must have 5V. The 12V you mention are the output ratings for the HUB's power supply. To be compatible with USB there must be an additional DCDC converter inside the HUB. Whether this HUB is capable of driving 2 RPis or not depends on the current it can provide, not its output voltage. A voltage of ~5.5V must not be exceeded! Note that ...


3

I doubt there is a definite answer to that. This lens is an OEM device and to my knowledge the Foundation's documentation does not provide any details about the lens, it's holder and their mechanical interface. Why am I asked to move to a different standard? I would contribute that to the availability of standard lenses of a different size, e.g. M12. ...


3

The source for the bcm2835-v4l2 kernel module is in the linux staging tree at linux/drivers/staging/vc04_services/bcm2835-camera/ You can find it e.g. in the RPF Linux repository at https://github.com/raspberrypi/linux For the current RPF Linux kernel (4.14.y) it's at https://github.com/raspberrypi/linux/tree/rpi-4.14.y/drivers/staging/vc04_services/...


3

Creating a name based on the current time is a common solution to this problem. As long as you know that you will not create two files within the same second, this is fine. Depending on your needs, you can format the time in different ways. If you only care about a unique name, using the time in seconds is fine. Sometimes it is desirable to have that time ...


2

Regarding the official answer on Raspberry Pi forum, it is not possible to include B-Frames into h264 stream since hardware encoder doesn't support it.


2

You should look at computer vision and the OpenCV library. To put it quickly, this problem has been solved before, and all you need to do is to google "recognizing numbers with opencv" to find tutorials. PyImageSearch is always a good resource for OpenCV. There are also many examples on YouTube. This is an example with license plates. OpenCV is one of the ...


2

How about something like the following. I'm using a list to store your pan and tilt values and capturing 1 image on each loop. Simply concatenating the string value of i onto 'image' followed by .'jpg' to create the image name. #!/usr/bin/env python3 import pantilthat,time,picamera panTilt = [[ 30, -10], [ 20, 0], [ 10, 10], ...


2

I've tried with last RASPBIAN STRETCH LITE (9.4 - 2018-06-27) in a PI3B v1.2 with a PI CAMERA 2.1, all fully upgraded (apt-get upgrade) If this command doesn't detect the camera with "detected=1" like this: pi@raspberrypi:~ $ vcgencmd get_camera supported=1 detected=1 Then there is a connection problem. Check if the cable both ways are with the good side ...


2

In the simplest form, the sender simply reads the file and writes its contents to the UART, while the receiver opens the UART and keeps reading bytes until there's nothing left (with a reasonable timeout), then stores these bytes to a file. For more complex cases, check out serial data transmission protocols such as XMODEM / YMODEM / ZMODEM. The advantage ...


1

I have the same issue. Raspberry Pi 3 B+, Raspbian Stretch 9, camera module v1.3, saving to SD. Using following flags, should be 12 pictures over 1 hour. raspistill -v -bm -vf -hf -awb cloud -br 60 -sa 40 -t 3600000 -tl 300000 -q 100 -o image%04d.jpg I was getting the issue where mmal skips frames and names an image randomly (or by time stamp?) and then ...


1

I have found that the solution mentioned in https://www.linux-projects.org/uv4l/installation/ works well. Out-of-the box, the video will have a text overlay, but if you give a "contribution" of $15 or so, and email the author, you can get a key that removes the text. I'm getting low-latency (0.5 sec or so) video at 1280 X 960 @ 30fps from a Model 3B ...


1

I tried excluding GL related shared libraries at linking to trace the function dependencies and was surprised brcmGLESv2 brcmEGL and vcos were not even needed for linking! In fact vcos was the one that broke Qt GL dependency. So leaving out brcmGLESv2 brcmEGL and vcos allows Qt GUI to start up and I can show the raspicam preview "over" a qt dialog window ...


1

The only way to play H264 video streams with standard web technologies is WebRTC. UV4L has been supporting WebRTC for years now, and the project website is plenty of examples, tutorials and demos made for the Raspberry Pi showing how to run custom web applications with UV4L itself providing H264 hardware-encoded streams in the pages.


1

have you had a look at motioneyeos. It's a pre-built distro for doing just that. You get a variety of options for controling the camera framerate etc or you can choose to have it as a simple IP camera and view the feed as you wish on any computer.


1

You can’t combine 2 composite video outputs - you would just get a corrupted output. You would need some form of switcher to switch one or the other.


1

Thanks everyone for suggestions. Finally I selected IVport multiplexer: http://www.ivmech.com/magaza/en/development-modules-c-4/ivport-raspberry-pi-camera-module-multiplexer-p-90 It allows up to 4 cameras per adapter, and by stacking 4 of them, up to 16 cameras per board. It is not very cheap though, but as far as I am concerned, reduces the amount of ...


1

You can record as it is like any other computers do, you just need to have a free space left on you sdcard todo so. Also, there is a life expectancy for every storage device, it is called the read and write cycle, it may not affect your sdcard directly but it may in a longer run. See: https://superuser.com/questions/17350/whats-the-life-expectancy-of-an-sd-...


1

You can save your video files directly to the SD card - provided you have sufficient space on the card. "Sufficient space" will obviously depend upon what size card you have, and how big your video files are. I assume you can calculate this, and you haven't provided file or SD card sizes, so... If you're worried about SD card corruption, you can always ...


1

You can turn on the camera led through the pi’s GPIO using the python code below: #!/usr/bin/env python import time import RPi.GPIO as GPIO # Use GPIO numbering GPIO.setmode(GPIO.BCM)   # Set GPIO for camera LED # Use 5 for Model A/B and 32 for Model B+ CAMLED = 5    # Set GPIO to output GPIO.setup(CAMLED, GPIO.OUT, initial=False)  GPIO....


1

Undefined reference errors have gone when include the vc_vchi_gencmd.h file with extern "C"... declaration; extern "C" { #include "interface/vmcs_host/vc_vchi_gencmd.h" }


1

Normally camera output will not show over VNC. It bypasses the framebuffer so VNC cannot capture it. If you have a screen directly connected to the Pi you would see the camera output on that. RealVNC has an experimental feature that will capture the camera output. It works, but it's not great. https://www.realvnc.com/en/connect/docs/raspberry-pi.html ...


1

You can get the camera status using vcgencmd get_camera on the command line. It will show if the camera is enabled / supported and if it is connected. pi@raspi05:$ vcgencmd get_camera supported=1 detected=0


1

I was getting VIDIOC_STREAMON: Operation not permitted while using cv2.VideoCapture(0). Setting the GPU memory to a higher value, solved the problem.


1

Install motion from github and build it yourself. This is how to do it: sudo mkdir motion sudo cd motion sudo apt-get install autoconf automake build-essential pkgconf libtool libzip-dev libjpeg-dev git libavformat-dev libavcodec-dev libavutil-dev libswscale-dev libavdevice-dev libwebp-dev gettext libmicrohttpd-dev git clone https://github.com/Motion-...


1

In Python you can use https://github.com/OsmoSystems/picamraw/ to extract the raw Bayer data or a rudimentary demosaiced RGB image. The result is a numpy array that you can use with a variety of other python tools. This basic example should display the raw image: First, install matplotlib and picamraw with pip install matplotlib picamraw. Then, import the ...


1

Have you tried capturing in separate scripts? You could write a new script for each resolution you wanted, store the files and have them managed or run via a processing script. Will try and build this myself over the weekend and see if I can add some code to this answer, but the theory should work.


1

Instead of using uv4l's raspicam driver to driver the raspberry pi camera, you can use the kernel-based bcm2835-v4l2 driver. Just modprobe it and enable it using raspi-config, and you should have a /dev/video0 file. After the device file appears, you can start uv4l with these options: uv4l --external-driver --device-name=video0 The text overlay will be ...


1

According to the Picamera Documentation in part 3.5 of the basic recipes section: You may wish to capture a sequence of images all of which look the same in terms of brightness, color, and contrast (this can be useful in timelapse photography, for example). Various attributes need to be used in order to ensure consistency across multiple shots. ...


1

After searching a lot I finally have this working now to get it working I used the following script for the configfs setup https://gist.github.com/kbingham/c39c4cc7c20882a104c08df5206e2f9f and this version of the uvc-gadget tool https://github.com/wlhe/uvc-gadget


1

I had the same problem. My Pi Zero W worked fine for 10 hours or so and then various kernel panic messages. The errors would get more frequent and at a point I could no longer connect to the Pi. It looks like the Pi is under-volted for this job. After some investigation I solved this by adding the following lines to the file /boot/config.txt over_voltage=...


Only top voted, non community-wiki answers of a minimum length are eligible