1:22 Gesture Controls with Teachable Machine Model 2:16 Image Classification 3:12 Object Detection 4:18 Google Coral USB Accelerator 4:52 Human Following 5:55 Object Tracking 6:50 Voice Control 7:25 Mobile Phone Accelerometer Control
Awesome project, sir. I made it without the 12v circuit and leds. I also used an L293d IC on a breadboard. Everything is powered by 5v usb power bank with two ports. All other functions with camera and ultrasonic sensor work fantastic. Thank a lot.
How would i be able to detect only specific class using your object detection functionality. Which section of the code should i modify? for instance how would i ignore the detection(display of the bounding box) of cellphone maybe?
There are 90 objects which the object detection ML model can detect. If any one is detected, bounding boxes are created around them by function "append_text_img1()" in "common1.py file". There, you can add a line "label = labels.get(obj.id, obj.id)" inside the for loop to get the label name of individual object. Now, you can decide whether to keep a particular lablel or reject it.
Respected Sir, Respected Sir, First of all your tutorial is awesome. It is working good. But when I try to replace mobilenet_ssd_v2_coco_quant_postprocess.tflite with my custom trained model for fire detection, it is not working. Could you kindly assist me in resolving this problem? Your guidance is crucial to me.
It's amazing that I found what I expected. Great job! If you can improve the wheels with Mecanum Wheels, it will be perfect!!! Anyway, thank you for your great work!
Hi Jitesh. Great work!! Thanks for posting. I've followed your tutorials and it is working ok on raspberry pi 4b, unless the Speaker, I can't make it work! Searching a lot here to find the problem I found the $.post is empty array in speaker.js to Ajax.php file. Only the text, but mp3 horn files is working.
Hi Arthur, glad you liked it. I guess you are facing issues with TTS functionality. I developed the GUI for it sometime back which ran perfectly on Stretch version of raspberry Pi OS. Later with Buster, I also faced issues with TTS functionality through web ui. In the backend I am using espeak module for TTS which works fine through terminal. Sounds are produced using omxplayer which had no issues with old or the current os. You can try installing the previous stretch version and try the same if TTS is absolutely essential for you
@@jitesh_helloworld Hi Jitesh tanks for your time. Today I made many tests even using other tts libraries in python, but it seems only work if I force the script running in terminal like you said, or debugging the .php script, but there's no response to tts requesting by index.php. Difficult to understand, because the speaker_tts.py is running ok, LEDs blinking when press the button on index.php Control Panel page.
Wow, this is amazing! Unfortunately I had numerous errors with the tensorflow lite, apparently the bash file provided gave some errors in the installation. Anyway, congratulations!
Thank you. Open the bash file, you will find a section marked tensorflow lite.. Run those commands manually.. I got inputs from others like you that the script ran successfully. Try it with fresh OS and uninterrupted internet connection. It should work
Nice channel to learn about robotics , can you also explore on ROS(Robot Operating System) too ? Possible also provide tutorial guide on how do u build the robot i'm interested to make one. Thanks alot ~
Thanks.. not ventured into ROS, may be in future.. If you want to build this robot, just assemble a basic chassis with DC motors and a driver board and a Pi Camera. This much is enough to get started. Read this article helloworld.co.in/article/basic-robotics-make-robot-raspberry-pi-web-controls Also, install the robot software on your Raspberry Pi from here helloworld.co.in/earthrover/ This is an automated bash script which installs every thing you need to run the code of this robot
Hi Jitesh, is there any way to connect with you? I'm working on a project and needed some help. Charge us for your time but if you could help us that would be really kind and great.
Hello friend, I am replicating your robot and I would like to expand it, connecting the raspberry pi with arduino, would you recommend me to use I2C for this purpose, or do you think it would be better through serial communication? I'm talking to you from Ecuador, you really made a great project, thanks for sharing your knowledge with the world. For robotics enthusiasts, people like you and projects like these help us a lot.
This just needs better response times and I can finally make my wasp zapper pulsed laser gun to kill wasps mid-flight as they circle around my food and I can't whack them with my slipper because I might hit my food.
1:22 Gesture Controls with Teachable Machine Model
2:16 Image Classification
3:12 Object Detection
4:18 Google Coral USB Accelerator
4:52 Human Following
5:55 Object Tracking
6:50 Voice Control
7:25 Mobile Phone Accelerometer Control
First, thanks a ton for sharing this, it’ll be my first project to learn Raspberry and stuff. Second, do you have a list of hardwares to build this?
@@cuongdinh2659 great. All the best
How can I communicate with you
@@vip8man829 jitesh@helloworld.co.in
Awesome project, sir. I made it without the 12v circuit and leds. I also used an L293d IC on a breadboard. Everything is powered by 5v usb power bank with two ports. All other functions with camera and ultrasonic sensor work fantastic. Thank a lot.
Glad it worked for you
Grandioso, gracias por compartir tus conocimientos me ha ayudado demasiado en mis proyectos de robotica.
Thanks
Waowww...Thats amazing , great work, i recently started with raspberry pi and esps and this channel just makes me more motivated, :)
Thankyou.. All the best with your exploration.. Raspberry Pi is pure fun
ottimo lavoro bellissimo.
Everything works perfect but the AI section is not working.There is no streaming of video in every section
Really great work!!
Thanks a lot
Can this work even if CORAL USB is not included/interface in the project?
5:00 to 5:55 is nice
Thanks
Could you tell me how to install and configure tensorflow lite manually? The bash file gave some errors = /
Great job , is rasberry pi 3b sufficient for all of these .
Is rpi 3b sufficient?
Yes it is sufficient
Can you please share the detail videos of how to make this full robot?
helloworld.co.in/article/basic-robotics-make-robot-raspberry-pi-web-controls
helloworld.co.in/article/camera-robot-using-raspberry-pi-web-controlled-surveillance-robot
How would i be able to detect only specific class using your object detection functionality. Which section of the code should i modify? for instance how would i ignore the detection(display of the bounding box) of cellphone maybe?
There are 90 objects which the object detection ML model can detect. If any one is detected, bounding boxes are created around them by function "append_text_img1()" in "common1.py file". There, you can add a line "label = labels.get(obj.id, obj.id)" inside the for loop to get the label name of individual object. Now, you can decide whether to keep a particular lablel or reject it.
@@jitesh_helloworld thanks for your reply. i tried this modification but didnt work what went wrong?
for obj in objs:
label =labels.get(obj.id, obj.id)
x0, y0, x1, y1 = list(obj.bbox)
x0, y0, x1, y1 = int(x0*width), int(y0*height), int(x1*width), int(y1*height)
percent = int(100 * obj.score)
if label == "person" or "dog" or "cat" or "bird" or "horse" or "elephant" or "zebra" or "bear":
if (percent>=60):
box_color, text_color, thickness=(0,255,0), (0,255,0),2
elif (percent40):
box_color, text_color, thickness=(0,0,255), (0,0,255),2
else:
box_color, text_color, thickness=(255,0,0), (255,0,0),1
text3 = '{}% {}'.format(percent, label.get(obj.id, obj.id))
cv2_im = cv2.rectangle(cv2_im, (x0, y0), (x1, y1), box_color, thickness)
cv2_im = cv2.putText(cv2_im, text3, (x0, y1-5),font, 0.5, text_color, thickness)
the camera live feed freezes , it would be great if you show me the code.
Respected Sir,
Respected Sir,
First of all your tutorial is awesome. It is working good. But when I try to replace mobilenet_ssd_v2_coco_quant_postprocess.tflite with my custom trained model for fire detection, it is not working. Could you kindly assist me in resolving this problem? Your guidance is crucial to me.
It's amazing that I found what I expected. Great job! If you can improve the wheels with Mecanum Wheels, it will be perfect!!! Anyway, thank you for your great work!
Thank you :)
Hi Jitesh. Great work!! Thanks for posting. I've followed your tutorials and it is working ok on raspberry pi 4b, unless the Speaker, I can't make it work! Searching a lot here to find the problem I found the $.post is empty array in speaker.js to Ajax.php file. Only the text, but mp3 horn files is working.
Hi Arthur, glad you liked it. I guess you are facing issues with TTS functionality. I developed the GUI for it sometime back which ran perfectly on Stretch version of raspberry Pi OS. Later with Buster, I also faced issues with TTS functionality through web ui. In the backend I am using espeak module for TTS which works fine through terminal. Sounds are produced using omxplayer which had no issues with old or the current os. You can try installing the previous stretch version and try the same if TTS is absolutely essential for you
@@jitesh_helloworld Hi Jitesh tanks for your time. Today I made many tests even using other tts libraries in python, but it seems only work if I force the script running in terminal like you said, or debugging the .php script, but there's no response to tts requesting by index.php. Difficult to understand, because the speaker_tts.py is running ok, LEDs blinking when press the button on index.php Control Panel page.
great project!
does it work on raspberry pi zero 2w?
Thanks. Not tried, but it should run
@@jitesh_helloworld OK, thanks a lot and best wishes again for this project!
Wow, this is amazing! Unfortunately I had numerous errors with the tensorflow lite, apparently the bash file provided gave some errors in the installation. Anyway, congratulations!
Thank you. Open the bash file, you will find a section marked tensorflow lite.. Run those commands manually.. I got inputs from others like you that the script ran successfully. Try it with fresh OS and uninterrupted internet connection. It should work
Nice channel to learn about robotics , can you also explore on ROS(Robot Operating System) too ? Possible also provide tutorial guide on how do u build the robot i'm interested to make one. Thanks alot ~
Thanks.. not ventured into ROS, may be in future.. If you want to build this robot, just assemble a basic chassis with DC motors and a driver board and a Pi Camera. This much is enough to get started. Read this article helloworld.co.in/article/basic-robotics-make-robot-raspberry-pi-web-controls
Also, install the robot software on your Raspberry Pi from here helloworld.co.in/earthrover/
This is an automated bash script which installs every thing you need to run the code of this robot
Hi Jitesh, is there any way to connect with you? I'm working on a project and needed some help. Charge us for your time but if you could help us that would be really kind and great.
Hi Shikhar. What help do you want. Describe your project. jitesh@helloworld.co.in
Hello friend, I am replicating your robot and I would like to expand it, connecting the raspberry pi with arduino, would you recommend me to use I2C for this purpose, or do you think it would be better through serial communication?
I'm talking to you from Ecuador, you really made a great project, thanks for sharing your knowledge with the world.
For robotics enthusiasts, people like you and projects like these help us a lot.
Serial Communication is a better choice. Let me know what you create . Thanks
👍👍👍👍
Since when are you interested in Robotics 😂 🤣
This just needs better response times and I can finally make my wasp zapper pulsed laser gun to kill wasps mid-flight as they circle around my food and I can't whack them with my slipper because I might hit my food.
Do send pics/video once you do that 👍
Hi
ottimo lavoro bellissimo