Escogitare

Raspberry Pi: i pin di GPIO

Sun, 02 Jul 2023 14:00:00 +0200

1: 3.3 V			2: 5 V
3: GPIO 2			4: 5 V
5: GPIO 3			6: massa
7: GPIO 4			8: UART trasmissione
9: massa			10: UART ricezione
11: GPIO 17			12: GPIO 18
13: GPIO 27			14: massa
15: GPIO 22			16: GPIO 23
17: 3.3 V			18: GPIO 24
19: GPIO 10			20: massa
21: GPIO 9			22: GPIO 25
23: GPIO 11			24: GPIO 8
25: massa			26: GPIO 7
27: riservato			28: riservato
29: GPIO 5			30: massa
31: GPIO 6			32: GPIO 12
33: GPIO 13			34: massa
35: GPIO 19			36: GPIO 16
37: GPIO 26			38: GPIO 20
39: massa			40: GPIO 21

Riferimenti

Raspberry Pi 4 Model B Datasheet: datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf

Scattare foto con Picamera 2 e Raspberry PI usando le coroutine in Python

Mon, 27 Feb 2023 13:00:00 +0100

Usare la libreria Picamera2 e Camera Module 3 è uno dei tanti modi per scattare foto co Raspberry PI. Il modo più semplice per usarla da programma è in Python, usando la libreria Picamera2. Una volta installato il modulo si può provare il codice proposto dalla documentazione ufficiale (The Picamera 2 Manual, cap 2.3):

from picamera2 import Picamera2, Preview import time picam2 = Picamera2() camera_config = picam2.create_preview_configuration() picam2.configure(camera_config) picam2.start_preview(Preview.QTGL) picam2.start() time.sleep(2) picam2.capture_file("test.jpg")

Questo programma avvia una finestra di preview, quindi in una board Raspberry Pi 4 headless (senza monitor e tastiera) può essere semplificato:

from picamera2 import Picamera2 cam = Picamera2() cam.start() cam.capture_file("test_1.jpg")

Se il programma è complesso e le elaborazioni sulle foto richiedono tempo, si può trarre vantaggio dall’esecuzione concorrente sui core del processore della scheda Raspberry. Python offre la libreria asyncio (Asynchronous I/O) che permette l’uso della sintassi async/await.

Quando il programma deve svolgere un compito gravoso, lo lancia in parallelo, aspettando la fine dell’esecuzione solo quando avrà bisogno dei risultati, come in figura:

Nell’esempio seguente, uso solo le api di avvio e sincronizzazione delle coroutines: le funzioni concorrenti vengono dichiarate come async e vengono lanciate in parallelo con asyncio.create_task(). Questo metodo restituisce un oggetto Task, che poi può essere usato per una sincronizzazione successiva usando il comando await:

#!/usr/bin/python3 import asyncio from picamera2 import Picamera2 async def shot(camera): # … some more code …  camera.capture_file("test_2.jpg") # … some more code …  async def main(): cam = Picamera2() cam.start() task1 = asyncio.create_task(shot(cam)) # … some more code …  await task1 asyncio.run(main())

Riferimenti

The Picamera 2 Manual: datasheets.raspberrypi.com/camera/picamera2-manual.pdf
Raspberry Pi Camera Module 3 : datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B Datasheet: datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Python asyncio, Asynchronous I/O: docs.python.org/3/library/asyncio.html

Prendre des photos avec Picamera 2 et Raspberry Pi à l'aide de coroutines en Python

Mon, 27 Feb 2023 13:00:00 +0100

L’utilisation de Picamera2 et du Camera Module 3 est l’une des nombreuses façons de prendre des photos avec Raspberry PI. Une fois installé le module hardware, vous pouvez essayer le code proposé par la documentation officielle (The Picamera 2 Manual, cap 2.3) :

from picamera2 import Picamera2, Preview import time picam2 = Picamera2() camera_config = picam2.create_preview_configuration() picam2.configure(camera_config) picam2.start_preview(Preview.QTGL) picam2.start() time.sleep(2) picam2.capture_file("test.jpg")

Ce programme lance une fenêtre de prévisualisation, donc dans une carte Raspberry Pi 4 headless (sans écran ni clavier) cela peut être simplifié :

from picamera2 import Picamera2 cam = Picamera2() cam.start() cam.capture_file("test_1.jpg")

Si le programme est complexe et que le traitement des photos prend du temps, vous pouvez profiter d’une exécution simultanée sur les cores du processeur de la carte Raspberry. Python propose asyncio (Asynchronous I/O) qui permet l’utilisation de la syntaxe async/await.

Lorsque le programme doit effectuer une tâche lourde, il la lance en parallèle, n’attendant la fin de l’exécution que lorsqu’il a besoin des résultats, comme dans la figure suivante :

Dans l’exemple suivant, j’utilise uniquement les functions de démarrage et de synchronisation des coroutines : les fonctions concurrentes sont déclarées comme async et sont lancées en parallèle avec asyncio.create_task(). Cette méthode renvoie un objet Task qui peut ensuite être utilisé pour une synchronisation ultérieure à l’aide de la commande await :

#!/usr/bin/python3 import asyncio from picamera2 import Picamera2 async def shot(camera): # … some more code …  camera.capture_file("test_2.jpg") # … some more code …  async def main(): cam = Picamera2() cam.start() task1 = asyncio.create_task(shot(cam)) # … some more code …  await task1 asyncio.run(main())

Références

The Picamera 2 Manual : datasheets.raspberrypi.com/camera/picamera2-manual.pdf
Raspberry Pi Camera Module 3 : datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B Datasheet : datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Python asyncio, Asynchronous I/O : docs.python.org/3/library/asyncio.html

Taking pictures with Picamera 2 and Raspberry Pi using coroutines in Python

Mon, 27 Feb 2023 13:00:00 +0100

Using the Picamera2 library and Camera Module 3 is one of the many ways to take pictures with Raspberry PI. Once the hardware module is installed you can try the code proposed by the official documentation (The Picamera 2 Manual, cap 2.3):

from picamera2 import Picamera2, Preview import time picam2 = Picamera2() camera_config = picam2.create_preview_configuration() picam2.configure(camera_config) picam2.start_preview(Preview.QTGL) picam2.start() time.sleep(2) picam2.capture_file("test.jpg")

This program launches a preview window, so in a headless Raspberry Pi 4 board (without monitor and keyboard) it can be simplified:

from picamera2 import Picamera2 cam = Picamera2() cam.start() cam.capture_file("test_1.jpg")

If the program is complex and photo processing takes time, you can take advantage of concurrent execution on the processor cores of the Raspberry board. Python offers the asyncio library (Asynchronous I/O) which allows the use of the async/await syntax.

When the program has to carry out some heavy tasks, it launches them in parallel, waiting for the end of the execution only when it needs the results, as in the figure:

In the following example, I only use the coroutines startup and sync apis: concurrent functions are declared as async and are launched in parallel with asyncio.create_task(). This method returns a Task object, which can then be used for a later sync using the await command:

#!/usr/bin/python3 import asyncio from picamera2 import Picamera2 async def shot(camera): # … some more code …  camera.capture_file("test_2.jpg") # … some more code …  async def main(): cam = Picamera2() cam.start() task1 = asyncio.create_task(shot(cam)) # … some more code …  await task1 asyncio.run(main())

Note that asyncio requires to launch the main function with asyncio.run(...).

References

The Picamera 2 Manual: datasheets.raspberrypi.com/camera/picamera2-manual.pdf
Raspberry Pi Camera Module 3 : datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B Datasheet: datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Python asyncio, Asynchronous I/O: docs.python.org/3/library/asyncio.html

Camera Module 3: come connetterla a Raspberry Pi 4

Mon, 20 Feb 2023 13:00:00 +0100

Camera Module 3 è la terza generazione dell’obiettivo fotografico per Raspberry Pi. Usa un sensore Sony IMX708 con autofocus e HDR. Queste sono le caratteristiche principali:

Risoluzione: 11.9 Megapixels
Dimensioni: 4608 x 2592 pixels
Profondità di campo minima: 10 cm (5 cm per la Camera Module 3 Wide)
Lunghezza focale: 4.74 mm (2.75 mm per la Camera Module 3 Wide)
Rapport focale: F1.8 (F2.2 per la Camera Module 3 Wide)
Tempo di esposizione massimo: 112 s

Camera Module 3 si collega al Raspberry Pi tramite il connettore CSI (Camera Serial Interface). Attenzione: la porta è molto simile a quella usata dal display, occorre fare attenzione alla posizione e alle scritte sulla scheda.

Per aprire il connettore è necessario afferrarlo ai lati più stretti e sollevarlo dolcemente. Quindi si inserisce il cavo con la parte blu rivolta verso la porta Ethernet in modo che i contatti metallici del cavo tocchino quelli del connettore.

Infine si deve chiudere il connettore spingendolo in basso.

Per verificare il funzionamento si possono usare le applicazioni libcamera-jpeg e libcamera-vid da shell, anche con una connessione ssh. Ad esempio libcamera-jpeg --list-cameras mostra le fotocamere installate:

pi@raspberrypi:~ $ libcamera-jpeg --list-cameras Available cameras ----------------- 0 : imx708 [4608x2592] (/base/soc/i2c0mux/i2c@1/imx708@1a) Modes: 'SRGGB10_CSI2P' : 1536x864 [120.13 fps - (0, 0)/4608x2592 crop] 2304x1296 [56.03 fps - (0, 0)/4608x2592 crop] 4608x2592 [14.35 fps - (0, 0)/4608x2592 crop]

Per catturare un’immagine invece si può usare il comando:

libcamera-jpeg -o test.jpg –immediate

La fotografia seguente mostra il risultato del comando: un selfie della scheda Raspberry Pi.

Riferimenti

Raspberry Pi Camera Module 3 : datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B Datasheet: datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Camera Serial Interface, pagina su Wikipedia: en.wikipedia.org/wiki/Camera_Serial_Interface

Camera Module 3: comment le connecter à Raspberry Pi 4

Mon, 20 Feb 2023 13:00:00 +0100

Camera Module 3 est la troisième génération de l’objectif de la caméra pour Raspberry Pi. Il utilise un capteur Sony IMX708 avec autofocus et HDR. Voici les principales caractéristiques :

Résolution : 11,9 mégapixels
Taille : 4 608 x 2 592 pixels
Profondeur de champ minimale : 10 cm (5 cm pour le Camera Module 3 Wide)
Distance focale : 4,74 mm (2.75 mm pour le Camera Module 3 Wide)
Rapport focal : F1.8 (F2.2 pour le Camera Module 3 Wide)
Temps d’exposition maximal : 112 s

La Camera Module 3 se connecte au Raspberry Pi via le connecteur CSI (Cameral Serial Interface). Attention : le port est très similaire à celui utilisé par l’écran, il faut faire attention à la position et à l’écriture sur la carte.

Pour ouvrir le connecteur, il est nécessaire de le saisir par les côtés les plus étroits, et de le soulever doucement. Ensuite, vous insérez le câble avec le côté bleu pointant vers le port Ethernet.

Enfin, le connecteur doit être fermé en le poussant doucement vers le bas.

Vous pouvez utiliser les applications libcamera-jpeg et libcamera-vid pour vérifier si la camera fonctionne. Par exemple, libcamera-jpeg --list-cameras affiche les caméras installées :

pi@raspberrypi:~ $ libcamera-jpeg --list-cameras Available cameras ----------------- 0 : imx708 [4608x2592] (/base/soc/i2c0mux/i2c@1/imx708@1a) Modes: 'SRGGB10_CSI2P' : 1536x864 [120.13 fps - (0, 0)/4608x2592 crop] 2304x1296 [56.03 fps - (0, 0)/4608x2592 crop] 4608x2592 [14.35 fps - (0, 0)/4608x2592 crop]

Pour capturer une image, vous pouvez utiliser :

libcamera-jpeg -o test.jpg –immediate

La photo suivante montre le résultat : un selfie de la carte Raspberry Pi.

Références

Raspberry Pi Camera Module 3 : datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B : datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Camera Serial Interface : fr.wikipedia.org/wiki/Camera_Serial_Interface

Camera Module 3: how to connect to Raspberry Pi 4

Mon, 20 Feb 2023 13:00:00 +0100

“Camera Module 3” is the third generation of the camera lens for Raspberry Pi. It uses a Sony IMX708 sensor with autofocus and HDR. These are the main features:

Resolution: 11.9 Megapixels
Size in pixels: 4608 x 2592 pixels
Minimum depth of field: 10 cm (5 cm for the Camera Module 3 Wide)
Focal length: 4.74mm (2.75 mm for the Camera Module 3 Wide)
Focal Ratio: F1.8 (F2.2 for the Camera Module 3 Wide)
Maximum exposure time: 112 s

Camera Module 3 connects to the Raspberry Pi via the CSI (Cameral Serial Interface) connector. Warning: the port is very similar to the one used by the display, you need to pay attention to the position and the writing on the board.

To open the connector it is necessary to grasp it by the narrower sides and lift it gently. Then you insert the cable with the blue side pointing towards the Ethernet port.

Finally, the connector must be closed by pushing it down (gently).

You can use the libcamera-jpeg and libcamera-vid applications to check if the camera works. For example libcamera-jpeg --list-cameras shows the installed cameras:

pi@raspberrypi:~ $ libcamera-jpeg --list-cameras Available cameras ----------------- 0 : imx708 [4608x2592] (/base/soc/i2c0mux/i2c@1/imx708@1a) Modes: 'SRGGB10_CSI2P' : 1536x864 [120.13 fps - (0, 0)/4608x2592 crop] 2304x1296 [56.03 fps - (0, 0)/4608x2592 crop] 4608x2592 [14.35 fps - (0, 0)/4608x2592 crop]

To capture an image you can use the command:

libcamera-jpeg -o test.jpg –immediate

The following photograph shows the result of the command: a selfie of the Raspberry Pi board.

References

Raspberry Pi Camera Module 3: datasheets.raspberrypi.com/camera/camera-module-3-product-brief.pdf
Raspberry Pi 4 Model B: datasheets.raspberrypi.com/rpi4/raspberry-pi-4-datasheet.pdf
Camera Serial Interface: en.wikipedia.org/wiki/Camera_Serial_Interface

The ECW file format for Crossword

Mon, 01 Oct 2018 14:00:00 +0200

The ECW file format can store in portable way a crossword. It is an XML text file based on the following DTD:

<!DOCTYPE cw [ <! ELEMENT cw (title, author, across, down, l+, h*)> <!ATTLIST cw rows CDATA #REQUIRED columns CDATA #REQUIRED lang CDATA style (dense | sparse) > <! ELEMENT title (#PCDATA) > <! ELEMENT author (#PCDATA) > <! ELEMENT across (w+) > <! ELEMENT down (w+) > <! ELEMENT w (#PCDATA) > <!ATTLIST w r CDATA #REQUIRED c CDATA #REQUIRED w CDATA #REQUIRED > <! ELEMENT l EMPTY > <!ATTLIST l r CDATA #REQUIRED l CDATA #REQUIRED > <! ELEMENT h EMPTY > <!ATTLIST h r CDATA #REQUIRED l CDATA #REQUIRED > ]>

It stores some metadata information as the author and the title, the words and the clues, and two different views of the schema (the solved schema an the optional hints).

An example could be:

<cw rows="9" columns="9" lang="en" style="dense"> <title>AutoCrossword easy 1</title> <author>Automatically generated from Wiktionary</author> <!-- Clues --> <across> <w r="0" c="0" w="KEV">kiloelectronvolt, 1000 eV</w> <w r="0" c="6" w="LDP">Linux Documentation Project</w> <w r="1" c="6" w="SOT">stupid person; fool</w> <w r="2" c="2" w="AMOSITE">A variety of grunerite asbestos, once used in insulation.</w> <w r="4" c="0" w="RACQ">Royal Automobile Club of Queensland (Australian Automobile Association)</w> <w r="4" c="5" w="PERT">&apos;P&apos;rogram &apos;E&apos;valuation and &apos;R&apos;eview &apos;T&apos;echnique, a method for diagramming and analyzing the flow of dependent tasks and other events in a project.</w> <w r="6" c="0" w="BEEFISH">resembling beef</w> <w r="7" c="0" w="MMS">Multimedia Messaging Service</w> <w r="8" c="0" w="DES">Data Encryption Standard</w> <w r="8" c="6" w="XMP">chemistry xanthosine monophosphate|</w> </across> <down> <w r="0" c="0" w="KDD">knowledge discovery in databases|</w> <w r="0" c="2" w="VIA">A main road or highway, especially in ancient Rome. &apos;(Mainly used in set phrases, below.)&apos;</w> <w r="0" c="4" w="RHO">Any of a family of GTP-binding proteins</w> <w r="0" c="6" w="LSI">large scale integration|</w> <w r="0" c="7" w="DOTARD">An old person with impaired intellect; one in his or her dotage.</w> <w r="0" c="8" w="PTE">Abbreviation of private (soldier).</w> <w r="2" c="3" w="MAQAF">The Hebrew script hyphen (mul|sc=Hebr|־)</w> <w r="2" c="5" w="SUPES">Superman, the &apos;DC Comics&apos; character.</w> <w r="3" c="1" w="RACEME">An indeterminate inflorescence in which the flowers are arranged along a single central axis.</w> <w r="6" c="0" w="BMD">Bermudian dollar</w> <w r="6" c="2" w="ESS">electronic switching system|electronic switching system|</w> <w r="6" c="4" w="IRQ">The ISO 3166-1 three-letter (alpha-3) code for Iraq.</w> <w r="6" c="6" w="HMX">An explosive, related to RDX, &apos;octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine&apos;</w> <w r="6" c="8" w="STP">The ISO 3166-1 three-letter (alpha-3) code for Sao Tome and Principe.</w> </down> <!-- Solution --> <l r="0" l="KEV*R*LDP" /> <l r="1" l="D*I*H*SOT" /> <l r="2" l="D*AMOSITE" /> <l r="3" l="*R*A*U*A*" /> <l r="4" l="RACQ*PERT" /> <l r="5" l="*C*A*E*D*" /> <l r="6" l="BEEFISH*S" /> <l r="7" l="MMS*R*M*T" /> <l r="8" l="DES*Q*XMP" /> <!-- Hints --> <h r="0" l=" * * " /> <h r="1" l=" * * * " /> <h r="2" l=" * " /> <h r="3" l="* * * * *" /> <h r="4" l=" * " /> <h r="5" l="* * * * *" /> <h r="6" l=" * " /> <h r="7" l=" * * * " /> <h r="8" l=" * * " /> </cw>

Using ffmpeg to convert from MTS to mp4

Sat, 22 Oct 2016 14:00:00 +0200

I have a bunch of memory cards packed with MTS files, that I recorded in few years. Now I would like to convert them in mp4, in order to watch them on my TV or iPad. I used the awesome ffpmeg to concatenate and convert them.

1. Create a text file with filenames

First I created a text file with the names of all origin videos. Every line in the file is like this:

file 'subdirectory/filename.MTS'

I just used the ls command (ls AVCHD/BDMV/STREAM/*MTS > files.txt) and few substitutions in a text editor to create it.

2. Conatenate

The fastest command is just a concatenation:

ffmpeg -f concat -i files.txt -c copy movie.mts

But you may want to create video files for DVDs, or save space using an H.264 encoding. For DVDs a command could be:

ffmpeg -f concat -i files.txt -target pal-dvd movie.mpg

For an mp4 file:

ffmpeg -f concat -i files.txt -c:a copy -c:v libx264 -preset slow -crf 18 -f mp4 movie.mp4 

You may tune the -preset and -crf parameters to have better quality, or faster encoding.

3. Do not disturb, please

Converting a video is a tough job: everything will slow down while doing. So, if you use your computer at the same time, you may want to put the conversion on a lower priority, using the command renice.

First you need to know the PID of the ffmpeg process with:

ps –al

Then you change its priority with renice:

renice -n 19 PID

Where ‘19’ is the lowest priority: it will takes a little more, but it will doesn’t bug you too much.

For example:

 ps –al ... renice –n 19 654

4. What about an iPad or Android tablet?

To optimize your videos for iPads, ffmpeg gives you some more flags:

ffmpeg -f concat -i files.txt -vcodec libx264 -profile:v main -level 3.1 -preset medium -tune film -x264-params ref=4 -acodec aac -ac 2 -strict experimental -movflags +faststart -f mp4 movie.mp4 

It looks like that family videos may benefice of -tune film parameter, while the -level depends on your iPad model. You may also need to scale down your video with: -vf "scale=-2:720:force_original_aspect_ratio=decrease".

More information in ffmpeg.org web site, or on Wikipedia.

Android Support Preferences Library: nested PreferenceScreen

Wed, 07 Sep 2016 14:00:00 +0200

Android Support Library includes two librares to simplify the implementation of settings pages:

A post from Ian Lake on Google+ shows how to use them. It is easy, but when you need to use multiple nested settings pages you need to pay attention to a couple of points:

You need to implement the onCreatePreferences method:

 @Override public void onCreatePreferences(Bundle savedInstanceState, String rootKey) { setPreferencesFromResource(R.xml.preferences, rootKey); }

Every PreferenceScreen in your preferences XML file must have a key:

 <PreferenceScreen android:key="prefscrkey">

This key will not be used in any setting, but in the onCreatePreferences, in order to understand the screen to show.