* Create dig-cont_0640_s3_q.tflite * Update edit_reference.html (#2924) Zoom-Einstellungen ausblenden, wenn zoom deaktiviert ist * Update edit_config_template.html (#2925) Zoom-Einstellungen ausblenden, wenn zoom deaktiviert ist + Kameraeinstellungen nicht nur in der config.ini speichern, sondern auch setzen * The parameter "negative" is processed on the MCU * Renamed menu entry Alignment -> References * renamed model file to make naming consistent * Fix test (#2933) * always reset change absolute (#2956) * Create dig-class11_1701_s2.tflite * Update tflite * Update tflite * Camera settings (#3029) * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Fix minor typo and harmonize spaces (#3030) * Update edit_digits.html * Update edit_analog.html * Update overview.html * Update edit_reference.html --------- Co-authored-by: jomjol <30766535+jomjol@users.noreply.github.com> * Source code formatting, ClassFlowPostProcessing.cpp * fix edit_config_template.html Unfortunately, I didn't take out some entries from other experiments because I overlooked/forgot them^^ * fix readconfigparam.js Unfortunately, I didn't take out some entries from other experiments because I overlooked/forgot them^^ * fix edit_config_template.html TakeImage_CamZoomMode_value1 was replaced by TakeImage_CamZoomSize_value1, but not deleted. This caused problems when saving the camera settings. * Update info.html function loadWLANSSID() was present three times * Update overview.html CamFrameSize is no longer needed/used for zoom * Update reply-bot.yaml * homeassistant service discovery: derive node_id when using nested topics (#3088) * derive correct node_id for homeassistant service discovery in nested topics (fixes #1792) * explicit use of std::string * move nodeId creation to separate function add unit-tests * add documentation about node_id generation for Home Assistant MQTT Service Discovery * add Prometheus/OpenMetrics exporter (#3081) * add prometheus endpoint * refine metrics implementation * move metrics generator to ClassFlowControll * add more metrics align prefix * add more metrics clean up * refine documentation * revert dependencies change * sanitize labels * create separate module for openmetrics * move openmetrics to separate folder * clean up * add basic unit-tests * work with const numbers add replaceAll for string replacement avoid opening std namespace adapt unit-tests * Update code/main/server_main.cpp --------- Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Typo * update platformIO to 6.7.0 (ESP IDF 5.2.1) (#3098) * update to platformio/espressif32 @ 6.7.0 * remove unused getReadout() as it throws errors (error: 'virtual std::string ClassFlow::getReadout()' was hidden). --------- Co-authored-by: CaCO3 <caco@ruinelli.ch> * Update reply-bot.yaml (#3107) * Fix actions-label-commenter * Update Helper.cpp * Update Helper.h * Update Helper.cpp * Update readconfigparam.js fix for: In the selected field the value '1' in the section 'TakeImage' in the field 'CamGainceiling' is invalid. PLEASE CHECK BEFORE SAVING! * proposal for renaming and documentation (#3115) * Update server_camera.cpp Fix building with -D DEBUG_DETAIL_ON better alternative to: https://github.com/jomjol/AI-on-the-edge-device/pull/3160 * Update ClassFlowPostProcessing.cpp * Update ClassFlowAlignment.cpp * Fix building with `-D DEBUG_DETAIL_ON` (#3160) Building with `-D DEBUG_DETAIL_ON` has been broken since #3029. Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Handle empty prevalue.ini gracefully (#3162) Fixes #2149. * Bugfix for boot loop (#3175) * Add files via upload * Add files via upload * Add files via upload * Delete param-docs/parameter-pages/TakeImage/Aec2.md has been replaced by CamAec2.md * Delete param-docs/parameter-pages/TakeImage/AutoExposureLevel.md has been replaced by CamAeLevel.md * Delete param-docs/parameter-pages/TakeImage/Brightness.md has been replaced by CamBrightness.md * Delete param-docs/parameter-pages/TakeImage/Contrast.md has been replaced by CamContrast.md * Delete param-docs/parameter-pages/TakeImage/Grayscale.md has been replaced by CamSpecialEffect.md * Delete param-docs/parameter-pages/TakeImage/Negative.md has been replaced by CamSpecialEffect.md * Delete param-docs/parameter-pages/TakeImage/Saturation.md has been replaced by CamSaturation.md * Delete param-docs/parameter-pages/TakeImage/Sharpness.md has been replaced by CamSharpness.md * Delete param-docs/parameter-pages/TakeImage/ImageQuality.md has been replaced by CamQuality.md * Delete param-docs/parameter-pages/TakeImage/Zoom.md has been replaced by CamZoom.md * Delete param-docs/parameter-pages/TakeImage/ZoomMode.md has been replaced by CamZoomSize.md * Delete param-docs/parameter-pages/TakeImage/ZoomOffsetX.md has been replaced by CamZoomOffsetX.md * Delete param-docs/parameter-pages/TakeImage/ZoomOffsetY.md has been replaced by CamZoomOffsetY.md * Delete param-docs/parameter-pages/TakeImage/ImageSize.md has been replaced by CamZoomSize.md * Delete param-docs/parameter-pages/TakeImage/FixedExposure.md has been replaced by CamAec.md * Delete param-docs/parameter-pages/Alignment/FlipImageSize.md has been replaced by CamVflip.md * Delete param-docs/parameter-pages/Alignment/InitialMirror.md has been replaced by CamHmirror.md * CamParameter documentation update https://github.com/jomjol/AI-on-the-edge-device/issues/3185 * typo * add Webhook #3148 (#3163) * WIP add Webhook * fix config html for webhook add tooltips for webhook * webhook: fix not enabling webhook * send webhook as json * Update ApiKey.md * webhook: fix only sending last "Number" * webhook JSON is now closer to the data log in CSV format * webhook: drop timeStampTimeUTC and switch from timeStamp to lastvalue like lokal csv to fix no timestamp on error --------- Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Bugfix for time stamp (#3180) * Update ClassFlowPostProcessing.cpp * Update ClassFlowDefineTypes.h * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update interface_webhook.cpp * Update readconfigcommon.js fix for: The same message("Image Contrast got enhanced") came up with "Update Marker" and "Enhance Image Contrast". * fix svg favicon The svg one got added in33893eb566but does not work on Firefox * Update platformIO to 6.8.1 (Contains ESP IDF 5.3) (#3196) * Update platformIO to 6.8.1 (ESP IDF 5.3) * removed now redundant typedef * updated IDF manifest hash * Add files via upload so it should work now * Update server_main.cpp --------- Co-authored-by: CaCO3 <caco@ruinelli.ch> Co-authored-by: michael <Heinrich-Tuning@web.de> * Update MainFlowControl.cpp * Add support for OV5640 camera (#3063) * Add support for OV5640 camera * clean up sharpness handling * limit sharpness range to -2 and +2 * refactor * Fix OV3660 sharpness handling * refactor sharpness handling * fix OV3660 zoom mode * reinstate aspect ratio via imageSize * Changed OV5640 full frame size to match datasheet * various fixes * add denoise config and general clean up * fix line endings to LF * Support enabling red blue swap via web interface * update jpeg quality limits * remove color swap config; color swap workaround dependent on vflip * fix missing commit * fix gain ceiling * Update cam vflip param page * fix typo: camdenoise, not camsdenoise * fix compile errors * Update MainFlowControl.cpp * Add rate threshold parameter (#3195) * still needs to be tested https://github.com/jomjol/AI-on-the-edge-device/issues/3143 * Update ClassFlowPostProcessing.cpp code formatting * Update ClassFlowDefineTypes.h code formatting * Update ClassFlowPostProcessing.h code formatting * Update edit_config_template.html * fix * Update config.ini * Update edit_config_template.html * Updated param doc * Rename parameters * Update edit_config_template.html * Update NUMBER.ChangeRateThreshold.md * Update NUMBER.ChangeRateThreshold.md --------- Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Update main.cpp * Update config.ini Adjusted camera settings to make the image brighter. * Update readconfigparam.js Adjusted camera settings to make the image brighter. * add optional ImageUpload for Webhook (#3174) * WIP add Webhook * fix config html for webhook add tooltips for webhook * webhook: fix not enabling webhook * send webhook as json * Update ApiKey.md * webhook: fix only sending last "Number" * webhook JSON is now closer to the data log in CSV format * webhook: add img upload * webhoop added config for imgupload * webhook html fixes * webhook: drop timeStampTimeUTC and switch from timeStamp to lastvalue like lokal csv to fix no timestamp on error * add checkbox for Webhook_UploadImg * Update sd-card/html/edit_config_template.html * Update edit_config_template.html * Update edit_config_template.html * Update edit_config_template.html * added a long timestamp to both webhook requests --------- Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Add files via upload (#3207) * Update ClassFlowPostProcessing.cpp deleted some unnecessary double entries RateType renamed to MaxRateType * Update ClassFlowDefineTypes.h RateType renamed to MaxRateType * Update text on recognition page * Update digital CNN * add a Delay between the WiFi reconnections (#3068) * add a Delay between the WiFi reconnections * log the delay between the WiFi reconnections move the delay after the log * cleanup * cleanup * cleanup * Update edit_alignment.html * cleanup * cleanup * Update platformIO to 6.9.0 (Contains ESP IDF 5.3.1) * Handle crash on corrupted model (#3220) * Upgrade esp-tflite-micro to 1.3.1 * Added log message to hint in case it crashes on loading a corrupted model --------- Co-authored-by: CaCO3 <caco@ruinelli.ch> * new dig-class100-173-s2-q on 23.800 images (#3257) * new dig-class100-173-s2-q on 23.800 images * platformio/espressif32 @ 6.8.1 for esp32cam-dev * Revert "platformio/espressif32 @ 6.8.1 for esp32cam-dev" This reverts commitcc9297d483. * not using platformio 6.1.16 * Revert "not using platformio 6.1.16" This reverts commitef18e4fae7. * moved number edit box styles into new file edit_style.css (#3262) changed input[type=number] from 60px to 3em, to show 3 digits with current font size. * added note about only TLS 1.2 is supported (#3213) * Renamed digital to digit (#3219) * renamed Digital to Digit * added param migration * Update .github/label-commenter-config.yaml * renamed AnalogDigitTransition* to AnalogToDigitTransition* --------- Co-authored-by: CaCO3 <caco@ruinelli.ch> * Update Changelog.md * Update Changelog.md * Prepare next release (#3267) (#3270) * Create dig-cont_0640_s3_q.tflite * Update edit_reference.html (#2924) Zoom-Einstellungen ausblenden, wenn zoom deaktiviert ist * Update edit_config_template.html (#2925) Zoom-Einstellungen ausblenden, wenn zoom deaktiviert ist + Kameraeinstellungen nicht nur in der config.ini speichern, sondern auch setzen * The parameter "negative" is processed on the MCU * Renamed menu entry Alignment -> References * renamed model file to make naming consistent * Fix test (#2933) * always reset change absolute (#2956) * Create dig-class11_1701_s2.tflite * Update tflite * Update tflite * Camera settings (#3029) * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Add files via upload * Fix minor typo and harmonize spaces (#3030) * Update edit_digits.html * Update edit_analog.html * Update overview.html * Update edit_reference.html --------- * Source code formatting, ClassFlowPostProcessing.cpp * fix edit_config_template.html Unfortunately, I didn't take out some entries from other experiments because I overlooked/forgot them^^ * fix readconfigparam.js Unfortunately, I didn't take out some entries from other experiments because I overlooked/forgot them^^ * fix edit_config_template.html TakeImage_CamZoomMode_value1 was replaced by TakeImage_CamZoomSize_value1, but not deleted. This caused problems when saving the camera settings. * Update info.html function loadWLANSSID() was present three times * Update overview.html CamFrameSize is no longer needed/used for zoom * Update reply-bot.yaml * homeassistant service discovery: derive node_id when using nested topics (#3088) * derive correct node_id for homeassistant service discovery in nested topics (fixes #1792) * explicit use of std::string * move nodeId creation to separate function add unit-tests * add documentation about node_id generation for Home Assistant MQTT Service Discovery * add Prometheus/OpenMetrics exporter (#3081) * add prometheus endpoint * refine metrics implementation * move metrics generator to ClassFlowControll * add more metrics align prefix * add more metrics clean up * refine documentation * revert dependencies change * sanitize labels * create separate module for openmetrics * move openmetrics to separate folder * clean up * add basic unit-tests * work with const numbers add replaceAll for string replacement avoid opening std namespace adapt unit-tests * Update code/main/server_main.cpp --------- * Typo * update platformIO to 6.7.0 (ESP IDF 5.2.1) (#3098) * update to platformio/espressif32 @ 6.7.0 * remove unused getReadout() as it throws errors (error: 'virtual std::string ClassFlow::getReadout()' was hidden). --------- * Update reply-bot.yaml (#3107) * Fix actions-label-commenter * Update Helper.cpp * Update Helper.h * Update Helper.cpp * Update readconfigparam.js fix for: In the selected field the value '1' in the section 'TakeImage' in the field 'CamGainceiling' is invalid. PLEASE CHECK BEFORE SAVING! * proposal for renaming and documentation (#3115) * Update server_camera.cpp Fix building with -D DEBUG_DETAIL_ON better alternative to: https://github.com/jomjol/AI-on-the-edge-device/pull/3160 * Update ClassFlowPostProcessing.cpp * Update ClassFlowAlignment.cpp * Fix building with `-D DEBUG_DETAIL_ON` (#3160) Building with `-D DEBUG_DETAIL_ON` has been broken since #3029. * Handle empty prevalue.ini gracefully (#3162) Fixes #2149. * Bugfix for boot loop (#3175) * Add files via upload * Add files via upload * Add files via upload * Delete param-docs/parameter-pages/TakeImage/Aec2.md has been replaced by CamAec2.md * Delete param-docs/parameter-pages/TakeImage/AutoExposureLevel.md has been replaced by CamAeLevel.md * Delete param-docs/parameter-pages/TakeImage/Brightness.md has been replaced by CamBrightness.md * Delete param-docs/parameter-pages/TakeImage/Contrast.md has been replaced by CamContrast.md * Delete param-docs/parameter-pages/TakeImage/Grayscale.md has been replaced by CamSpecialEffect.md * Delete param-docs/parameter-pages/TakeImage/Negative.md has been replaced by CamSpecialEffect.md * Delete param-docs/parameter-pages/TakeImage/Saturation.md has been replaced by CamSaturation.md * Delete param-docs/parameter-pages/TakeImage/Sharpness.md has been replaced by CamSharpness.md * Delete param-docs/parameter-pages/TakeImage/ImageQuality.md has been replaced by CamQuality.md * Delete param-docs/parameter-pages/TakeImage/Zoom.md has been replaced by CamZoom.md * Delete param-docs/parameter-pages/TakeImage/ZoomMode.md has been replaced by CamZoomSize.md * Delete param-docs/parameter-pages/TakeImage/ZoomOffsetX.md has been replaced by CamZoomOffsetX.md * Delete param-docs/parameter-pages/TakeImage/ZoomOffsetY.md has been replaced by CamZoomOffsetY.md * Delete param-docs/parameter-pages/TakeImage/ImageSize.md has been replaced by CamZoomSize.md * Delete param-docs/parameter-pages/TakeImage/FixedExposure.md has been replaced by CamAec.md * Delete param-docs/parameter-pages/Alignment/FlipImageSize.md has been replaced by CamVflip.md * Delete param-docs/parameter-pages/Alignment/InitialMirror.md has been replaced by CamHmirror.md * CamParameter documentation update https://github.com/jomjol/AI-on-the-edge-device/issues/3185 * typo * add Webhook #3148 (#3163) * WIP add Webhook * fix config html for webhook add tooltips for webhook * webhook: fix not enabling webhook * send webhook as json * Update ApiKey.md * webhook: fix only sending last "Number" * webhook JSON is now closer to the data log in CSV format * webhook: drop timeStampTimeUTC and switch from timeStamp to lastvalue like lokal csv to fix no timestamp on error --------- * Bugfix for time stamp (#3180) * Update ClassFlowPostProcessing.cpp * Update ClassFlowDefineTypes.h * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update ClassFlowPostProcessing.cpp * Update interface_webhook.cpp * Update readconfigcommon.js fix for: The same message("Image Contrast got enhanced") came up with "Update Marker" and "Enhance Image Contrast". * fix svg favicon The svg one got added in33893eb566but does not work on Firefox * Update platformIO to 6.8.1 (Contains ESP IDF 5.3) (#3196) * Update platformIO to 6.8.1 (ESP IDF 5.3) * removed now redundant typedef * updated IDF manifest hash * Add files via upload so it should work now * Update server_main.cpp --------- * Update MainFlowControl.cpp * Add support for OV5640 camera (#3063) * Add support for OV5640 camera * clean up sharpness handling * limit sharpness range to -2 and +2 * refactor * Fix OV3660 sharpness handling * refactor sharpness handling * fix OV3660 zoom mode * reinstate aspect ratio via imageSize * Changed OV5640 full frame size to match datasheet * various fixes * add denoise config and general clean up * fix line endings to LF * Support enabling red blue swap via web interface * update jpeg quality limits * remove color swap config; color swap workaround dependent on vflip * fix missing commit * fix gain ceiling * Update cam vflip param page * fix typo: camdenoise, not camsdenoise * fix compile errors * Update MainFlowControl.cpp * Add rate threshold parameter (#3195) * still needs to be tested https://github.com/jomjol/AI-on-the-edge-device/issues/3143 * Update ClassFlowPostProcessing.cpp code formatting * Update ClassFlowDefineTypes.h code formatting * Update ClassFlowPostProcessing.h code formatting * Update edit_config_template.html * fix * Update config.ini * Update edit_config_template.html * Updated param doc * Rename parameters * Update edit_config_template.html * Update NUMBER.ChangeRateThreshold.md * Update NUMBER.ChangeRateThreshold.md --------- * Update main.cpp * Update config.ini Adjusted camera settings to make the image brighter. * Update readconfigparam.js Adjusted camera settings to make the image brighter. * add optional ImageUpload for Webhook (#3174) * WIP add Webhook * fix config html for webhook add tooltips for webhook * webhook: fix not enabling webhook * send webhook as json * Update ApiKey.md * webhook: fix only sending last "Number" * webhook JSON is now closer to the data log in CSV format * webhook: add img upload * webhoop added config for imgupload * webhook html fixes * webhook: drop timeStampTimeUTC and switch from timeStamp to lastvalue like lokal csv to fix no timestamp on error * add checkbox for Webhook_UploadImg * Update sd-card/html/edit_config_template.html * Update edit_config_template.html * Update edit_config_template.html * Update edit_config_template.html * added a long timestamp to both webhook requests --------- * Add files via upload (#3207) * Update ClassFlowPostProcessing.cpp deleted some unnecessary double entries RateType renamed to MaxRateType * Update ClassFlowDefineTypes.h RateType renamed to MaxRateType * Update text on recognition page * Update digital CNN * add a Delay between the WiFi reconnections (#3068) * add a Delay between the WiFi reconnections * log the delay between the WiFi reconnections move the delay after the log * cleanup * cleanup * cleanup * Update edit_alignment.html * cleanup * cleanup * Update platformIO to 6.9.0 (Contains ESP IDF 5.3.1) * Handle crash on corrupted model (#3220) * Upgrade esp-tflite-micro to 1.3.1 * Added log message to hint in case it crashes on loading a corrupted model --------- * new dig-class100-173-s2-q on 23.800 images (#3257) * new dig-class100-173-s2-q on 23.800 images * platformio/espressif32 @ 6.8.1 for esp32cam-dev * Revert "platformio/espressif32 @ 6.8.1 for esp32cam-dev" This reverts commitcc9297d483. * not using platformio 6.1.16 * Revert "not using platformio 6.1.16" This reverts commitef18e4fae7. * moved number edit box styles into new file edit_style.css (#3262) changed input[type=number] from 60px to 3em, to show 3 digits with current font size. * added note about only TLS 1.2 is supported (#3213) * Renamed digital to digit (#3219) * renamed Digital to Digit * added param migration * Update .github/label-commenter-config.yaml * renamed AnalogDigitTransition* to AnalogToDigitTransition* --------- * Update Changelog.md * Update Changelog.md --------- Co-authored-by: jomjol <30766535+jomjol@users.noreply.github.com> Co-authored-by: michael <Heinrich-Tuning@web.de> Co-authored-by: Frank Haverland <fspapaping@googlemail.com> Co-authored-by: kub3let <95883234+kub3let@users.noreply.github.com> Co-authored-by: Marco H <myxor@users.noreply.github.com> Co-authored-by: Henry Thasler <henrythasler@users.noreply.github.com> Co-authored-by: CaCO3 <caco@ruinelli.ch> Co-authored-by: Sebastian Lövdahl <slovdahl@hibox.fi> Co-authored-by: Raphael Hehl <raphael@rhehl.de> Co-authored-by: jasaw <jasaw@dius.com.au> Co-authored-by: Francesco Carnielli <hex7c0@gmail.com> Co-authored-by: kalwados <kalwados@gmx.de> * revert TFlite submodule update as certain modules crash with it, see https://github.com/jomjol/AI-on-the-edge-device/issues/3269 * Rename NUMBER.AnalogDigitTransitionStart.md to NUMBER.AnalogToDigitTransitionStart.md * Update NUMBER.AnalogToDigitTransitionStart.md * Update edit_config_template.html * changed the webhook UploadImg to false * changed the webhook UploadImg to false (#3279) Co-authored-by: CaCO3 <caco@ruinelli.ch> * Show boolean option in settings page (#3285) * changed the webhook UploadImg to false * show boolean value on boolean drop down boxes --------- Co-authored-by: CaCO3 <caco@ruinelli.ch> * changed default value from boolean to numeric value in parameter camDenoise documentation * Update edit_explain_7.html https://github.com/jomjol/AI-on-the-edge-device/issues/3284?notification_referrer_id=NT_kwDOAAj1wbIxMjY3MDI1Njk3MDo1ODcyMDE#issuecomment-2387102786 * docs: update README.md (#3290) prefered -> preferred * rename/remove unused parameters (#3291) * migrate-cam-parameters (#3288) * Update main.cpp * remove commented out code --------- Co-authored-by: CaCO3 <caco3@ruinelli.ch> * Update Changelog.md * cleanup * Skip release jobs on non-release changes * Update build.yaml * Update build.yaml * Update build.yaml * Update build.yaml * Update build.yaml * fix pipeline * Update build.yaml --------- Co-authored-by: jomjol <30766535+jomjol@users.noreply.github.com> Co-authored-by: michael <Heinrich-Tuning@web.de> Co-authored-by: Frank Haverland <fspapaping@googlemail.com> Co-authored-by: kub3let <95883234+kub3let@users.noreply.github.com> Co-authored-by: Marco H <myxor@users.noreply.github.com> Co-authored-by: Henry Thasler <henrythasler@users.noreply.github.com> Co-authored-by: CaCO3 <caco@ruinelli.ch> Co-authored-by: Sebastian Lövdahl <slovdahl@hibox.fi> Co-authored-by: Raphael Hehl <raphael@rhehl.de> Co-authored-by: jasaw <jasaw@dius.com.au> Co-authored-by: Francesco Carnielli <hex7c0@gmail.com> Co-authored-by: kalwados <kalwados@gmx.de> Co-authored-by: Ikko Eltociear Ashimine <eltociear@gmail.com>
Welcome to the AI-on-the-edge-device
Artificial intelligence based systems have become established in our everyday lives. Just think of speech or image recognition. Most of the systems rely on either powerful processors or a direct connection to the cloud for doing the calculations there. With the increasing power of modern processors, the AI systems are coming closer to the end user – which is usually called edge computing. Here, this edge computing is put into a practically oriented example, where an AI network is implemented on an ESP32 device so: AI on the edge.
This project allows you to digitize your analog water, gas, power and other meters using cheap and easily available hardware.
All you need is an ESP32 board with a supported camera and something of a practical hand.
Key features
- Tensorflow Lite (TFlite) integration – including easy-to-use wrapper
- Inline image processing (feature detection, alignment, ROI extraction)
- Small and cheap device (3 x 4.5 x 2 cm³, < 10 EUR)
- Integrated camera and illumination
- Web interface for administration and control
- OTA interface for updating directly via the web interface
- Full integration into Homeassistant
- Support for Influx DB 1 and 2
- MQTT
- REST API
Workflow
The device takes a photo of your meter at a defined interval. It then extracts the Regions of Interest (ROIs) from the image and runs them through artificial intelligence. As a result, you get the digitized value of your meter.
There are several options for what to do with that value. Either send it to an MQTT broker, write it to an InfluxDb or simply provide access to it via a REST API.
Impressions
AI-on-the-edge-device on a Water Meter
Web Interface (Water Meter)
AI-on-the-edge-device on a Electrical Power Meter
Setup
There is growing documentation which provides you with a lot of information. Head there to get a start, set it up and configure it.
There are also articles in the German Heise magazine "make:" about the setup and technical background (behind a paywall): DIY - Setup
A lot of people created useful Youtube videos which might help you getting started. Here a small selection:
- youtube.com/watch?v=HKBofb1cnNc
- youtube.com/watch?v=yyf0ORNLCk4
- youtube.com/watch?v=XxmTubGek6M
- youtube.com/watch?v=mDIJEyElkAU
- youtube.com/watch?v=SssiPkyKVVs
- youtube.com/watch?v=MAHE_QyHZFQ
- youtube.com/watch?v=Uap_6bwtILQ
For further background information, head to Neural Networks, Training Neural Networks and Programming on the ESP32.
Download
The latest available version can be found on the Releases page.
Flashing the ESP32
Initially you will have to flash the ESP32 via a USB connection. Later updates are possible directly over the air (OTA using WIFI).
There are different ways to flash your ESP32:
- The preferred way is the Web Installer and Console which is a browser-based tool to flash the ESP32 and extract the log over USB:

- Flash Tool from Espressif
- ESPtool (command-line tool)
See the documentation for more information.
Flashing the SD Card
The SD card can be setup automatically after the firmware got installed. See the documentation for details. For this to work, the SD card must be FAT formated (which is the default on a new SD card). Alternatively the SD card still can be setup manually, see the documentation for details!
Casing
Various 3D-printable housing can be found here:
- https://www.thingiverse.com/thing:4573481 (Water Meter)
- https://www.thingiverse.com/thing:5028229 (Power Meter)
- https://www.thingiverse.com/thing:5224101 (Gas Meter)
- https://www.thingiverse.com/thing:4571627 (ESP32-cam housing only)
Donate
If you would like to support the developer with a cup of coffee, you can do that via PayPal.
Support
If you have any technical problems please search the discussions. In case you found a ug or have a feature request, please open an issue.
In other cases you can contact the developer via email: ![]()
Changes and History
See Changelog.
Build It Yourself
See Build Instructions.
Tools
- Logfile downloader and combiner (Thx to reserve85)
- Files see '/tools/logfile-tool', how-to see documentation
Additional Ideas
There are some ideas and feature requests which are not currently being pursued – mainly due to capacity reasons on the part of the developers. They features are collected in the issues and in FeatureRequest.md.



