@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
@@Circuitdigest Loved your content, I have tried myself the project, but when i try to compile and upload, it shows me "Error Compiling for board AI Thinker ESP32-cam"
@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
I'm just loving it.Looking for such informative video! Can you make another Video to show the output result on web instead of showing on the OLED? And please make another video on plant disease identification project like this. Wish you luck. Thanks a lot
Look closely at the serial monitor , I think the address for the web is being generated on the 9th line of the serial monitor , I think if we copy paste it on the search bar in Google like he did for the 1st code i guess it would work showing the detection on video with indicators.
@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
Great video! I like what you did there in the end with the OLED screen, so could you please show your final code? Cause Im having a little struggle with it. Thanks
Nice content. I follow all the same steps, but dont know why i am getting below error in verious line of codes. conv.cpp:1789:67: error: either all initializer clauses should be designated or none of them should be 1789 | .channels = input->dims->data[3], 1 | ^
I resolved the above issue Just editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/ Disabling #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN 1 and set it to 0 will allow the project to compile ."
Excellent video. Thank you. Few questions: 1) do you know if the libraries included in VScode? 2) Any idea how far can such a detection work with this camera?
You can use the same library with VScode and Platform IO. the detection range would be very small since the ESP32 can only process image with lower resolution. if you increase the resolution the frame rate will be reduced drastically and the ESP32 may even crash with very high resolutions.
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/: and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
I already made that in dec 2023 .and guys use esp32 cam to take pictures .some will tell you to use mobile or pc cam it will work but using esp32 cam will improve your AI model . and if you want to upgrade it try using RPi that thing is made for image processing . use zero if you are using it for first time
I am trying to do that using esp-idf, but is harder than that surely! At professinal application we need be capable to port the project in esp-idf, I am almost giving up
No actually you are training your own AI model . More pictures you provide more effective your system will work. Edge impulse is use to create library for your ai module. Other wise you have to code for it . Which is hard and complicated and more chance of failure. And in future if you want upgrade your system you can make new library and add to your code and upload it
Maker sure you have the latest esp32 board definition installed. And keep in mind that the board names are not in alphabetic order. If you are using Arduino IDE 2.0.x it would be easier if you search for the board using the select board dropdown menu in the main screen instead of using the tools menu.
Sir, this is my first time on this channel. Sir, I made a car detection project using Edge Impulse, but when I uploaded it to the ESP32-CAM, I encountered many initializer errors during compilation. Sir, please tell me why this happened when everything was auto-generated.
Sir , when i this video i didn't find anything you are writing thr code is getting itself by including the libraries and selecting that library. Is any need to write code or the library itself provides the code?
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp: In function 'TfLiteStatus tflite::{anonymous}::Prepare(TfLiteContext*, TfLiteNode*)': C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1789:67: error: either all initializer clauses should be designated or none of them should be 1789 | .channels = input->dims->data[3], 1 | ^ C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1793:68: error: either all initializer clauses should be designated or none of them should be 1793 | .channels = output->dims->data[3], 1 | ^ C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1795:80: error: either all initializer clauses should be designated or none of them should be 1795 | data_dims_t filter_dims = {.width = filter_width, .height = filter_height, 0, 0}; | ^ C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp: In function 'void tflite::{anonymous}::EvalQuantizedPerChannel(TfLiteContext*, TfLiteNode*, const TfLiteConvParams&, const NodeData&, const TfLiteEvalTensor*, const TfLiteEvalTensor*, const TfLiteEvalTensor*, TfLiteEvalTensor*)': C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1883:58: error: either all initializer clauses should be designated or none of them should be 1883 | .channels = input_depth, 1 | ^ C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1887:59: error: either all initializer clauses should be designated or none of them should be 1887 | .channels = output_depth, 1 | ^ C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1889:80: error: either all initializer clauses should be designated or none of them should be 1889 | data_dims_t filter_dims = {.width = filter_width, .height = filter_height, 0, 0}; | ^ exit status 1 Compilation error: exit status 1
Sir I am using ESP32-S camera module. It is not from AI thinker. Can you please suggest what changes should I make in the code because it is showing error and the code is not uploading
Most likely they would be the same. Check if the example camera server code is working or not. If not change the camera pin definitions, according to your dev board.
can you do sound detection on esp32 using I2s mems please. detect sound and distinguish it example if it is fire alarm dog bark something like commonly heard on our surroundings
i have uploaded zip library and did exactly as shown in video to run esp32cam code. its showing error before compilation finished. error is "The error you're encountering indicates issues with the use of initializer lists in your code. The error message either all initializer clauses should be designated or none of them should be points to improper mixing of designated and non-designated initializers." can you please resolve this. trying from 1 month.
Hi sir, how to integrate servomotors so that i can sort the object after detecting the object in the same code by edge impulse, how can i modify the code ?
Yes but be careful while supplying power to it but . Use FTDI it will save you from wiring headache or you can use ESP32-CAM programing board for 200rs
@@Giburozu I didn’t get any error while using this certain version 2.0.4. The project is running smoothly till now sir . I don’t have any idea why u get such error
Nice video❤ I want to on red led when onion is detected, green when potato is detected and yellow when tomato is detected, is it possible ? , if yes how can i do this ?
@@Circuitdigest it is a great solution but I not understand why the same model would be less ac curate on an esp32, slower I agree, but less accurate? Sure?
@@SA-oj3bo During training the model we reduced the training cycle so as to reduce the code size and make it easy for esp32. ideally if you need better accuracy you should provide more training datasets and use high epoch but this is not feasible on a small processor like esp32. also we cannot use a better resolution camera with esp32
Hi, I did the same thing and uploaded the zip file but I am getting error in compilation where it says to do sone changes in conv.cpp file and they are related to .channel and .output. What to do in that case?
Either all initializer clauses should be designated or none of them should be data_dims_t filter _dims={.width=filter__width,.height=filter_height,0,0}; how to correct this error.
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/: and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
14:27 It show me: Error: 13 INTERNAL: Library install failed: moving extracted archive to destination dir: copying library to destination directory:: creating destination dir
Either all initializer clauses should be designated or none of them should be data_dims_t filter _dims={.width=filter__width,.height=filter_height,0,0}; how to correct this error.pls guide me step by step because im the beginner of doing project this is my first project.pls help me sir..thanks in advance
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/: and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
The first compile after opening the IDE will always slower. You can enable the verbose output for compile in the preferences to see the compilations progress.
hi, may i know why when i uploaded the code it showed compilation error and also some In function 'TfLiteStatus tflite::{anonymous}::Prepare(TfLiteContext*, TfLiteNode*)': , either all initializer clauses should be designated or none of them should be 1789 | .channels = input->dims->data[3], 1
Make sure to follow steps exactly as in the article. If you still face the same issue try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/: Disabling #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN 1 and set it to 0 will allow the project to compile .
hello sir, I absolutely loved your video. I have a query, pls help me look into it. Ater saving impulse, when raw features are to be generated, it is saying "Error: VipsJpeg: Invalid SOS parameters for sequential JPEG" please help sir, this is causing a major setback in my projject.
@@CircuitdigestThankyou, I did find out that this is caused by samsung phones, as pics taken using samsung devices have some error data, resaving the image in another device or format will work.
make sure the Arduino is in reset state. and as I mentioned earlier use a board with 3.3V logic. And you must press and hold the boot / GPIO0 button while the ESPCAM board is connected to put it in the download mode.
Thanks for the video Actually I am facing difficulties regarding one project that I got demanded from school Actually the project is school involves four houses and hence there are four coloured tshirt so they just wanted to collect the data that how many students are present of red colour tshirt or blue color tshirt and so on. So kindly help me out. Or please make a video related to this project I would really appreciate you for this kind. Thank you so much sir 😊
Look are you doing this for first time . First don't directly jump to it .and for this I will recommend you Rpi zero for it and using rpi camera it is little bit expensive but will do the job. You will get rpi zero around 1200 rs to 1300rs depends from were you are buying this and 500 rs for camera but it's quality is better than esp32 cam .
Check your serial monitor it will give you an Ip address copy it and paste it to your browser. But I will recommend you. If you are using it for first time try do streaming with esp32cam . First it will help you in this project
Well you can use arduino if you want but use FTDI man ! 100 or 200 rs and if you don't want use it try using ESP32-CAM programing board same price but will save your lot of time
Bro after build the AI trained Model.the code is not Compiling it takes long time and they display "compilation terminated error". Please help me bro.to solve this problem.
check from which variable the detection is coming from, and use if condition to check the class. Finally, use digitalWrite() to perform HIGH & LOW operations. Thanks!
Glad to see your video made 10x my views in just 10 days T_T. That said, I'm happy that more people get to know my library
Hey thank you! your work is also amazing, great job!!!
I think its possible, not sure though
@@Circuitdigest hi sir,
Which algorithm is this working on
@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
@@Circuitdigest Loved your content, I have tried myself the project, but when i try to compile and upload, it shows me "Error Compiling for board AI Thinker ESP32-cam"
Yours is the only Indian Tech Channel worth watching, all others make such basic videos, but your projects has some standard.
best tutorial for beginners using Edge Impulse.
Thanks for the feedback! Glad you found it helpful.
sir i love you unlike other ytuber u explain everything
This is on another level
BRO SET VIDEO AYRUNU NJAN KORA THIVASAM AYEE ITHINTA PORAKA NADAKAN THUDANJITEE BUT YOU SAVED MY DAY
same avasthayanu aliya
very practical video, congrats.
Nice video 👍 gonna try it asap. Keep uploading informative videos like this.
Thank you, I will
@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
Thank you keep upgrading us
Excellent thank you.
Glad you found it useful!
Hocam tam aradığım video ❤
fantastic video sir working perfect
Very Very Thank you Sir
Glad you found it helpful.
If you encounter problem , on Windows just modify the tensorflow NN in library source
Could you tell me what change is made?
Wonderful sir, thank you!
I'm just loving it.Looking for such informative video!
Can you make another Video to show the output result on web instead of showing on the OLED?
And please make another video on plant disease identification project like this.
Wish you luck.
Thanks a lot
Look closely at the serial monitor , I think the address for the web is being generated on the 9th line of the serial monitor , I think if we copy paste it on the search bar in Google like he did for the 1st code i guess it would work showing the detection on video with indicators.
Nice video, but zoom in on the Arduino IDE text so it's easier to view on a mobile device. Thanks!
You can get all the necessary details on the project page linked in the video description.
@@Circuitdigest hello sir, is it not possible to have a bounding box or something similar? The point is to detect objects in real time with an ESP32 cam that has a bounding box or centeroid
@@Circuitdigest hello sir I have an error on the ai thinker board, what version of the esp32 board are you using?
Thaks u Sir, nice
Nice video, could you please show what you've modified in the code for the oled display? Thanks
Great topic, thanks 👍
Great video! I like what you did there in the end with the OLED screen, so could you please show your final code? Cause Im having a little struggle with it. Thanks
The final code is available in the project page. Please check it out.
very nice.. enjoyed
Thanks. Very nice job
can it be use to detect plastic bottles?
Yes with proper training data.
Nice content.
I follow all the same steps, but dont know why i am getting below error in verious line of codes.
conv.cpp:1789:67: error: either all initializer clauses should be designated or none of them should be
1789 | .channels = input->dims->data[3], 1
| ^
I resolved the above issue
Just editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/
Disabling #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN 1 and set it to 0 will allow the project to compile ."
@@rohitchaudhari5337 what to edit over there??
Thank you so much, you save me!
Excellent video. Thank you. Few questions: 1) do you know if the libraries included in VScode? 2) Any idea how far can such a detection work with this camera?
You can use the same library with VScode and Platform IO. the detection range would be very small since the ESP32 can only process image with lower resolution. if you increase the resolution the frame rate will be reduced drastically and the ESP32 may even crash with very high resolutions.
@@Circuitdigest hello sir I have an error on the ai thinker board, what version of the esp32 board are you using?
@@MuhammadRidho-qu9ey 3.0.1
'data_dims_t' {aka 'data_dims'} has no non-static data member named 'depth'
1850 | data_dims_t filter_dims = {.width = filter_width, .height = filter_height, .depth = 0, .batch = 0}; error occurred while uploading code
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/:
and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
@Circuitdigest thanks, sorted
I already made that in dec 2023 .and guys use esp32 cam to take pictures .some will tell you to use mobile or pc cam it will work but using esp32 cam will improve your AI model . and if you want to upgrade it try using RPi that thing is made for image processing . use zero if you are using it for first time
Very good explanation, but you didn't explain how to connect the OLED screen. What did you change in the code to display the results on the screen?
Please check the project page linked in the video description.
could you do a video with servos pan and tilt and esp32 cam (recognizing and tracking?)
If you want I can help you I made same thing a year ago
I am trying to do that using esp-idf, but is harder than that surely! At professinal application we need be capable to port the project in esp-idf, I am almost giving up
You can refer this docs.edgeimpulse.com/docs/tutorials/lifecycle-management/ota-model-updates/ota-model-updates-espressif-idf
Can you give sound output instead of text
You can with modified code with TTS.
yes you can do that remove the oled part from the code
Great
it is showing an error that the directory is not found even though I've installed it
Whats the exact error?
Is the full code in the link in bio?
as in, does the AI have the code or does the edge impulse have its own ai that it uses to train?
No actually you are training your own AI model . More pictures you provide more effective your system will work. Edge impulse is use to create library for your ai module. Other wise you have to code for it . Which is hard and complicated and more chance of failure. And in future if you want upgrade your system you can make new library and add to your code and upload it
Great project Sir. How many classifications that Esp-cam can store? Does your project have any restrictions?
Actually esp32 don't hold anything It's your library which you are creating more pictures you add to it more efficiently your system wil work
I want to calculate, the distance from the ESP32CAM to the object, How i can do this with the size of the centroid?
sir its not showing 'ai thinker esp32-cam' on arduino board
Maker sure you have the latest esp32 board definition installed. And keep in mind that the board names are not in alphabetic order. If you are using Arduino IDE 2.0.x it would be easier if you search for the board using the select board dropdown menu in the main screen instead of using the tools menu.
can i use edge impulse for training my esp 32 cam with the measured map of my home or map of any other place ? if yes, then please give the steps.
pls answer
What are you try to implement, using the training data?
YOLOOOOOOO!~ nice vid
Sir, this is my first time on this channel. Sir, I made a car detection project using Edge Impulse, but when I uploaded it to the ESP32-CAM, I encountered many initializer errors during compilation. Sir, please tell me why this happened when everything was auto-generated.
You will have to give a bit more context.
Can you do a video on how to debug an Edge/YOLO project using a JTAG debugger (ESP-Prog)?
Gald to see Indian IT guy working also on hardware
Sir, are you using esp32 cam in your sorting part video clip?
Which video?
In this video sir 0:20 to 0:26
Thats another project. But you can use the ESP32 CAM for the same purpose. Train the model according to your needed.
@@Circuitdigest OKY SIR, THANK U SO MUCH.
Have you guys noticed that "labeling queue" doesn't exist anymore? I am trying to find where to set the bounding boxes now.
Sir , when i this video i didn't find anything you are writing thr code is getting itself by including the libraries and selecting that library. Is any need to write code or the library itself provides the code?
Check the project page for detailed procedure.
Help, my code won't upload. It keeps saying compilation error: exit status 1
You need to be a bit more elaborate. First check what the error is. Without knowing the exact error we won't be able to help you.
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp: In function 'TfLiteStatus tflite::{anonymous}::Prepare(TfLiteContext*, TfLiteNode*)':
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1789:67: error: either all initializer clauses should be designated or none of them should be
1789 | .channels = input->dims->data[3], 1
| ^
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1793:68: error: either all initializer clauses should be designated or none of them should be
1793 | .channels = output->dims->data[3], 1
| ^
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1795:80: error: either all initializer clauses should be designated or none of them should be
1795 | data_dims_t filter_dims = {.width = filter_width, .height = filter_height, 0, 0};
| ^
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp: In function 'void tflite::{anonymous}::EvalQuantizedPerChannel(TfLiteContext*, TfLiteNode*, const TfLiteConvParams&, const NodeData&, const TfLiteEvalTensor*, const TfLiteEvalTensor*, const TfLiteEvalTensor*, TfLiteEvalTensor*)':
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1883:58: error: either all initializer clauses should be designated or none of them should be
1883 | .channels = input_depth, 1
| ^
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1887:59: error: either all initializer clauses should be designated or none of them should be
1887 | .channels = output_depth, 1
| ^
C:\Users\acer\Documents\Arduino\libraries\elinahcaralian-project-1_inferencing\src\edge-impulse-sdk\tensorflow\lite\micro\kernels\conv.cpp:1889:80: error: either all initializer clauses should be designated or none of them should be
1889 | data_dims_t filter_dims = {.width = filter_width, .height = filter_height, 0, 0};
| ^
exit status 1
Compilation error: exit status 1
@@Circuitdigest nevermind, I saw the fix you sent someone in a different comment, it works now
Sir I am using ESP32-S camera module. It is not from AI thinker.
Can you please suggest what changes should I make in the code because it is showing error and the code is not uploading
Most likely they would be the same. Check if the example camera server code is working or not. If not change the camera pin definitions, according to your dev board.
@@Circuitdigest please tell me how to change the camera pin definitions sir
I think mine is Wrover esp32 mudule
can you do sound detection on esp32 using I2s mems please. detect sound and distinguish it example if it is fire alarm dog bark something like commonly heard on our surroundings
yes
i have uploaded zip library and did exactly as shown in video to run esp32cam code. its showing error before compilation finished. error is "The error you're encountering indicates issues with the use of initializer lists in your code. The error message either all initializer clauses should be designated or none of them should be points to improper mixing of designated and non-designated initializers." can you please resolve this. trying from 1 month.
Just put the code to the chatgpt. It will give u a solution
And so when you pasted that into chat GPT what did it say?
nicely done! quick clarification - @5:39 GPIO0 should be held High or Low ?
GPIO should be pulled to LOW.
Can you provide source code for this
Hey bro can you tell me what software to design the prototype of this tutorial?
hi want to ask, my OLED is not working its not showing any characters just the bluee.... any recommendation from you?
Make sure your OLED is properly wired and also the i2C address is the same.
Hi, it’s the same way for the Xiao Esp32s3 ? Help please
Yes, you can use the same process with the ESP32-S3. Make sure the change the camera related codes accordingly.
Hi sir, how to integrate servomotors so that i can sort the object after detecting the object in the same code by edge impulse, how can i modify the code ?
You can use the servo motor library to control the servo after object detection. The modifications has to be done within the Arduino code.
Can I use Arduino uno to program the esp32 CAM
Yes but be careful while supplying power to it but . Use FTDI it will save you from wiring headache or you can use ESP32-CAM programing board for 200rs
can i detect text with this way, upload images to edge impluse and when esp32 cam detect this text on TV screen can recognize it ?
You can do character recognition. But it may need a lot of training data.
@Circuitdigest thanks
I tried two times now to download the file from edge impulse and both times I get an code 1 error. The file is not compiling correctly. Any ideas?
I found out that the latest package for Espressif - 3.0.7 was the problem. I downgraded the package to 2.0.7 and it worked.
Glad you got it fixed.
For the ESP32 board, what version of the library is used?
2.0.4
@@mr.emptygamingytwhen I use 2.0.4 version, why error no such file or directory
@@Giburozu I didn’t get any error while using this certain version 2.0.4. The project is running smoothly till now sir . I don’t have any idea why u get such error
@@mr.emptygamingyt huft.. Thanks
Nice video❤
I want to on red led when onion is detected, green when potato is detected and yellow when tomato is detected, is it possible ? , if yes how can i do this ?
Yes you can activate a corresponding LED when a specific object is detected. You can check for the bb.label value and activate the LEDs accordingly.
Could you use this to make the esp32CAM read barcodes?
Yes you can. Checkout our QR Code API.circuitdigest.com/microcontroller-projects/esp32-cam-qr-code-scanner
Will the accuracy with esp32 be lower or only slower compared to a solution on RP or PC? Which Yolo model is used for this ESP32 application?
The frame rate and accuracy will definitely better on RP or PC. We selected ESP32 cam because its a low cost option
@@Circuitdigest it is a great solution but I not understand why the same model would be less ac
curate on an esp32, slower I agree, but less accurate? Sure?
@@SA-oj3bo During training the model we reduced the training cycle so as to reduce the code size and make it easy for esp32. ideally if you need better accuracy you should provide more training datasets and use high epoch but this is not feasible on a small processor like esp32. also we cannot use a better resolution camera with esp32
Hello, could you please tell ne how much did it cost you overall for the products? Roughly!
Cost of the ESP32 camera module is around 3-6$.
Around 1300 rs for all of the components
How can the edge impulse web application itself tells it can identify objects with 100 percent accuracy without testing with an object?
It will use the test data set for evaluation.
Sir , can you make a programming video on this or share here?
If you need help I can help you
Hi, I did the same thing and uploaded the zip file but I am getting error in compilation where it says to do sone changes in conv.cpp file and they are related to .channel and .output. What to do in that case?
SUPISTE CORREGIRLO?
sir can i give a data set from google instead of live images
does that work
Yes. you can.
can we use 0.91 inch oled screen which is long..will it work?
plss tell
you can but you may need to change the text positions and all to fit it within the display if needed.
How to integrate in blynk? Labelling of objects by camera
Effect of ambient light on performance??
Since we are working with gray scale images lighting difference is tolerable. But if you want reliable results lighting should be maintained constant
Can you do on car plate
We have already done that. Here is the video. ruclips.net/video/c9bNmJvqJE4/видео.html
how to upgrade plan for edge impulse? I am not able to train the model for a large dataset
You will have to do that through the edge impulse website. The professional plan costs around 400$ per month.
Either all initializer clauses should be designated or none of them should be data_dims_t filter _dims={.width=filter__width,.height=filter_height,0,0}; how to correct this error.
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/:
and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
I didn't understand sir..I'm beginner of doing project..pls guide me step by step.. thanks in advance 🎉
Can we use this for facial recognition?
Yes. Here is the face recognition project. circuitdigest.com/microcontroller-projects/esp32-cam-face-recognition-using-edge-impulse
@@Circuitdigest thanks alot
Hii sir can you help me doing the same project like this for face recognition attendance system
use your face to create ai model instead of potato. and if you want some thing more advance use RPI
14:27
It show me:
Error: 13 INTERNAL: Library install failed: moving extracted archive to destination dir: copying library to destination directory:: creating destination dir
try manually extracting the library to Arduino libraries folder.
@Circuitdigest thank you, it works perfectly...
But can i edit the example with Arduino project like if the onion detected a green led works...
Yes you can.
Either all initializer clauses should be designated or none of them should be data_dims_t filter _dims={.width=filter__width,.height=filter_height,0,0}; how to correct this error.pls guide me step by step because im the beginner of doing project this is my first project.pls help me sir..thanks in advance
Try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/:
and set #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN from 1 to 0
in board manager what version of esp32 you use
2.x.x
can we use this to detect the number of cars in traffic ?, please reply
You can. you will need to train you model accordingly.
Can you help me? Why when im compiling, the process is so slow like dont do any progress?
The first compile after opening the IDE will always slower. You can enable the verbose output for compile in the preferences to see the compilations progress.
hi, may i know why when i uploaded the code it showed compilation error and also some In function 'TfLiteStatus tflite::{anonymous}::Prepare(TfLiteContext*, TfLiteNode*)': , either all initializer clauses should be designated or none of them should be
1789 | .channels = input->dims->data[3], 1
Make sure to follow steps exactly as in the article. If you still face the same issue try editing the file ei_classifier_config.h in exported Arduino library folder: /scr/edge-impulse-sdk/classifier/:
Disabling #define EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN 1 and set it to 0 will allow the project to compile .
@@Circuitdigest this works. thanks, I had same issue
@@6901ben it didnt work for me i did edit the file it still showed the same issue
pls elab
how did u get the coordinate data x,y,width,height at 15:44
It will be read from the detected bounding box properties.
It show me this :
upload error : failed uploading:no upload port provided copy error messages
Want can i do sir
Select the correct com port and make sure to put the board into download mode using the GPIO0 pin.
Which port we have to use whether the com or usb serial
The port that corresponds to the uart bridge.
hello sir, I absolutely loved your video. I have a query, pls help me look into it. Ater saving impulse, when raw features are to be generated, it is saying "Error: VipsJpeg: Invalid SOS parameters for sequential JPEG" please help sir, this is causing a major setback in my projject.
Either change your training images. Images taken with some phones may show this kind of error. Use the ESP32 CAM to take the training images.
@@CircuitdigestThankyou, I did find out that this is caused by samsung phones, as pics taken using samsung devices have some error data, resaving the image in another device or format will work.
Hi sir please suggest some good major project,I need it for my sem exam
If can make Drones include this systems to it with some modifications
Thanks
Can you program the robot
What robot?
Can't upload camera code I get error 'Compilation error: 'ledcSetup' was not declared in this scope; did you mean 'ledc_stop'?
make sure to install Eloquent ESP32-CAM Library.
hello can i use Arduino uno board instead of USB to Serial Converter
You will have to use the one with 3.3V logic.
@@Circuitdigest how to fix esp32 a fatal error occurred: failed to connect to esp32: no serial data received. i am using arduino uno
make sure the Arduino is in reset state. and as I mentioned earlier use a board with 3.3V logic. And you must press and hold the boot / GPIO0 button while the ESPCAM board is connected to put it in the download mode.
Thanks for the video
Actually I am facing difficulties regarding one project that I got demanded from school
Actually the project is school involves four houses and hence there are four coloured tshirt so they just wanted to collect the data that how many students are present of red colour tshirt or blue color tshirt and so on. So kindly help me out. Or please make a video related to this project I would really appreciate you for this kind.
Thank you so much sir 😊
Look are you doing this for first time . First don't directly jump to it .and for this I will recommend you Rpi zero for it and using rpi camera it is little bit expensive but will do the job. You will get rpi zero around 1200 rs to 1300rs depends from were you are buying this and 500 rs for camera but it's quality is better than esp32 cam .
@Vader141D yes I m doing this for first time
@@adityaarya-p9g man I am telling you it won't work as you want so better upgrade it to rpi esp32 cam is not for image processing
my serial monitor is not responding so what should i do?????
Make sure the connections are correct. and after uploading the code you might need to reset the board manually.
Sir, can we use BLE 33 SENSE as programmer?
i am using I2c lcd for display the result ,but ihave error on it can you help to display on I2c lcd
You will need to change the code accordingly. adapt the display related functions for the I2C LCD.
@Circuitdigest can you help me, im using ur code
@@Circuitdigest pls
Bro how to get stream link? I want video of the video stream link alone and not settings page
Check your serial monitor it will give you an Ip address copy it and paste it to your browser. But I will recommend you. If you are using it for first time try do streaming with esp32cam . First it will help you in this project
can i use OV7670? @
Circuit Digest
Its better to use OV2640 since the OV7670 lacks JPG mode.
Bro i just want to upload the code to the ESP-32 CAM through ESP 8266, so is there any other process need to be follow for that approach
Well you can use arduino if you want but use FTDI man ! 100 or 200 rs and if you don't want use it try using ESP32-CAM programing board same price but will save your lot of time
Is it work to detect human faces
With properly trained model yes.
Bro after build the AI trained Model.the code is not Compiling it takes long time and they display "compilation terminated error". Please help me bro.to solve this problem.
Nobody can help you if you don't provide the exact error code.
@@Circuitdigest c:\Users\Mohamed Mushraf\Documents\Arduino\libraries\Waste_Segregation_System-project-1_inferencing\src\edge-impulse-sdk\porting\espressif\ESP-NN\src\convolution\esp_nn_conv_s8_filter_aligned_input_padded_esp32s3.S:2: fatal error: opening dependency file C:\Users\Mohamed Mushraf\AppData\Local\Temp\arduino\sketches\D60EA5CD2FB35EF678691B1E576D39DC\libraries\Waste_Segregation_System-project-1_inferencing\edge-impulse-sdk\porting\espressif\ESP-NN\src\convolution\esp_nn_conv_s8_filter_aligned_input_padded_esp32s3.S.d: No such file or directory
2 | #if EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN && EI_CLASSIFIER_TFLITE_ENABLE_ESP_NN_S3
compilation terminated.
exit status 1
Compilation error: exit status 1
Why are you using ESP32 S3? which camera module are you using? For normal ESP32 camera module select the appropriate ESP32 board.
How to get an output (when detect union.output high or low)
check from which variable the detection is coming from, and use if condition to check the class. Finally, use digitalWrite() to perform HIGH & LOW operations. Thanks!
You can use the GPIO pins on ESP32CAM to get output based on what you detect