Did you enjoy this video? Try my premium courses! 😃🙌😊 ● Hands-On Computer Vision in the Cloud: Building an AWS-based Real Time Number Plate Recognition System bit.ly/3RXrE1Y ● End-To-End Computer Vision: Build and Deploy a Video Summarization API bit.ly/3tyQX0M ● Computer Vision on Edge: Real Time Number Plate Recognition on an Edge Device bit.ly/4dYodA7 ● Machine Learning Entrepreneur: How to start your entrepreneurial journey as a freelancer and content creator bit.ly/4bFLeaC Learn to create AI-based prototypes in the Computer Vision School! www.computervision.school 😃🚀🎓
Do you still reply to comments if so what is the program you are using to code? do you know if its available on raspberry PI 4? I'm new to all this so sorry if im being dumb and its common knowlage
Can we make something where camera detects the colour in traffic light and tells what colour is the light and it will assist the colourblind people if they are confused. Blinking yellow and blinking red is tough for them. It may be a good project.
i am facing the following error cv2.error: OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\imgproc\src\color.cpp:182: error: (-215:Assertion failed) !_src.empty() in function 'cv::cvtColor'
Thank you! It is so concise! May I know could this be further refined to detect objects based on the criteria like shape (eg rectangle in different orientation) and size (eg pixel x pixel)? Thanks!
Can you do a tutorial on how to regognise light a few colors light and the maybe log in into a excel for data to use. i want to make a uptime logger for my cnc machines
Is there a way to display an array at the center of the webcam and if a certain color gets in that array it will send some sort of signal to a mechanism?
I'm thinking of a device in the form of a video camera. This device will consist of a RaspberryPi, a camera module and display. This device will demonstrate the operation of the program for participation in roboracing. I propose to create a team to work on this project. In the future, it will be possible to participate in the OpenCV AI Competition 2023+n with this project. From Uruguay, you can participate in the competition)
Excuse me Sir, please can I ask you a question: I just decided to pay for the OpenCV certification, its 1500 canadian dollars, and I m pretty convinced that it is a great investment for my future career in CV. Please can you give me your opinion? It will be a great help Thank you in advance :)
Hey, yes it is possible, you could try by getting the bounding boxes of all connected components, the same way I did in the parking spot detector and counter video. 💪🙌
Hello, I'm trying to create a program that detects and counts objects by color, but now I can only count four colors, these are red, blue, green and yellow. I wonder if I can add the whole color scale to the program, how can I do this? I will be very happy if you give back. have a nice day😊
Hey, I think adding the entire color scale to the program will be very challenging. I would suggest to add more colors incrementally until performance is affected. 🙌
like it, really cool video.. the kicker for me was the "BRG" .. it really confused me for a while.. was experimenting with the values there.. was trying to read the color of a led which i thought was green but was more white:ish with a green halo haha, is it possible to detect more colors at the same time ?
Yeah, I hear you, something is tricky to specify the exact hue of a given color. About detecting more than one color at the same time: yes, absolutely! The code can be adapted to detect as many colors as you wish! 😃💪
Great video but I have a slight issue it says TypeError: VideoCapture() takes no arguments. I'm assuming this means its having trouble finding my webcam footage but I'm not sure how to fix it
Hey, I took a look and there was a bug which affected the performance in the case of red hues, thank you for noticing it. It is fixed now, take a look at the repo. 💪🙌
Hello Pasindu, I am not sure if I understand your question, what do you mean? In this video I show you how to detect colors using Python and OpenCV using the HSV colorspace.
As long as you can run Python, install OpenCV and get images from a webcam it 'should' work. If you try it on a raspberry pi let me know how it works! 💪
hey i tried to follow along with your tutorial but for some reason when i try to run it it says IndexError: index 10 is out of bounds for axis 0 with size 1 how can i fix this ?
Hey, define a new variable called red = [0, 0, 255] and use it instead of the 'yellow' variable. You may need to adjust the definition of red in case is not the exact hue of red you are looking for! 😃🙌
Hey Jerome, try defining a new variable called red = [0, 0, 255], and use that one instead of 'yellow'. If it doesn't detect the exact hue of red you are looking for you would have to see what are the bgr coordinates of the exact color you want to detect 💪🙌
Very useful tutorial, thanks! I would like to know if you could help me with a project. Do you offer paid consulting? The concept of the project is to create a system for selecting and separating stones by shade of green using a Raspberry Pi with a camera. The system captures real-time images through the camera and processes these images to detect the green tones in the stones. Based on the shades of green detected, the stones are classified into different groups (1, 2, 3, 4) according to the defined shade scale. The image generated by the camera is displayed on a monitor, and each stone is marked with a square that indicates which group it belongs to.
Hi, i need to make a robot that can avoid green and red obstacles. when the robot detect the green obstacle it should go left and when it sees red obstacle it should go right. and I am kinda lost how to make the camera program, can you help me?🥲
Hey, are you working with Arduino? If you are, search for videos on 'arduino color sensor project', there are plenty of videos on youtube which cover the foundations of how to do it. 🙌
How do you put a circle dot in the middle of the lemon/color? I'm trying to create a posture and gait analysis thesis with colored markers and get measurements (angles/distance), I want to be able to detect it and put a point in the middle of each markers and connect each point with lines. Hoping for your response, thanks in advance!
Hey, about the circle take a look at cv2.circle function, you can compute the coordinates of the center of the color by inspecting the bounding box coordinates. 🙌
Hey! Not sure if you still need the code but here it is anyway: if bbox is not None: x1, y1, x2, y2 = bbox cv2.rectangle(frame, (x1, y1), (x2, y2), [0, 255, 0], 5) center_x = (x1 + x2) // 2 center_y = (y1 + y2) // 2 frame = cv2.circle(frame, (center_x, center_y), 0, (0, 0, 255), 5) # change the 5 to adjust thickness just replace the if statement where you draw the rectangle and there should now me a circle in the middle of the object
Hey, is it possible to make it so the code runs the tracking on screen? Im trying to make an extension that runs on my pc on top of all my apps to track certain colors for premiere keyframes. Im not that skilled in coding so was hopping for a little help.
Hey Sergey, you want to track colors in your screen? Sure, it is possible, I guess the code in this project is a good baseline for that. Take a look at PyAutoGUI library, you can use it to take screenshots, for example. 💪💪
Is there a way to have multiple bounding boxes when there is more than 1 object of the specified color. Currently is makes 1 large bounding box, which inevitably covers more than the expected color?
Did you enjoy this video? Try my premium courses! 😃🙌😊
● Hands-On Computer Vision in the Cloud: Building an AWS-based Real Time Number Plate Recognition System bit.ly/3RXrE1Y
● End-To-End Computer Vision: Build and Deploy a Video Summarization API bit.ly/3tyQX0M
● Computer Vision on Edge: Real Time Number Plate Recognition on an Edge Device bit.ly/4dYodA7
● Machine Learning Entrepreneur: How to start your entrepreneurial journey as a freelancer and content creator bit.ly/4bFLeaC
Learn to create AI-based prototypes in the Computer Vision School! www.computervision.school 😃🚀🎓
I'm your big fan ,following since you have started channel. Now I'm unable to deploy my project live on github . Can you help me to live the project.
This video has just been the 1st and the best video I have landed for my CV project. Thank you for this video!
Glad you enjoyed it! I hope it helps with your project! 😃🙌
This is so perfectly what I was looking for. Thank you so much! Might enable me to build a great production inline measurement system for "a penny".
my favourite channel for CV projects, tysm as always!
Love this. Best course to get started.
This was such an amazing tutorila thank you so much!
Great video. Thank you for sharing your knowledge on computer vision
hey
i am getting error in this line
hsvImage = cv2.cvtColor(imageFrame, cv2.COLOR_BGR2HSV)
please tell me how to solve
cv2.cvtColor- in this
Excellent Sir, thank you! :)
Amazing channel, god bless you❤
Thank you! 😃🙌
תודה רבה 💪👏
😃🙌
this is better than pysource tutorial. how is it that all spanish guys are experts in CV? :)
Lol 😂
Cool explanation, thank you sir
You are welcome! Glad you enjoyed it! 😃🙌
Very useful tutorial, thanks.
You are welcome! 😃🙌
so cool.thank you, learned a lot!
😃🙌
Do you still reply to comments if so what is the program you are using to code? do you know if its available on raspberry PI 4? I'm new to all this so sorry if im being dumb and its common knowlage
Can we make something where camera detects the colour in traffic light and tells what colour is the light and it will assist the colourblind people if they are confused. Blinking yellow and blinking red is tough for them. It may be a good project.
Yeah it sounds like a good project idea! 🙌
i am facing the following error
cv2.error: OpenCV(4.6.0) D:\a\opencv-python\opencv-python\opencv\modules\imgproc\src\color.cpp:182: error: (-215:Assertion failed) !_src.empty() in function 'cv::cvtColor'
make sure you have selected the correct camera, if you only have one camera on your device then the correct line would be:
cap = cv2.VideoCapture(0)
Thank you! It is so concise! May I know could this be further refined to detect objects based on the criteria like shape (eg rectangle in different orientation) and size (eg pixel x pixel)? Thanks!
Yes, it can be done. You can further refine it to detect objects of a given color based on size and approx orientation. 🙌
Your code is not properly working tho
Can you do a tutorial on how to regognise light a few colors light and the maybe log in into a excel for data to use. i want to make a uptime logger for my cnc machines
How have you chosen a upper and lower limit? You have used some formula and I couldn't understand it?
The 'color' is given by the hue channel, so from the h value for yellow, I set the interval as +- 10.
Is there a way to display an array at the center of the webcam and if a certain color gets in that array it will send some sort of signal to a mechanism?
I'm thinking of a device in the form of a video camera. This device will consist of a RaspberryPi, a camera module and display. This device will demonstrate the operation of the program for participation in roboracing. I propose to create a team to work on this project. In the future, it will be possible to participate in the OpenCV AI Competition 2023+n with this project. From Uruguay, you can participate in the competition)
Sounds great! I think I won't be able to join, but best of luck with your project! 😃🙌
Very useful tutorial, thank you!
i would like to know how to make it work on black?
Have you tried by changing the bgr color to black (0, 0, 0)? Let me know if it works! 🙌
In the get limits function
c = np.uint8([[yellow]])
is throwing error
Excuse me Sir, please can I ask you a question: I just decided to pay for the OpenCV certification, its 1500 canadian dollars, and I m pretty convinced that it is a great investment for my future career in CV.
Please can you give me your opinion? It will be a great help
Thank you in advance :)
sir can u plz tell me where your running your code
Thannnk You!!!!!! That was smoooooooooothhhhhhhhhhh
sir, Does this affect the camera lighting?
How can I use my elgato capture card as the capture, I've tried looking everywhere?
Hi how can i connect the system to the alarm system specifically a buzzer while using a laptop camera for the video capturing, hope you will assist me
did you get it
The video is good, but I am not able to find the code for this tutorial. Could you provide the code link. Thanks
hello can you help me, i have some problem in here. i got message ModuleNotFoundError: No module named 'PIL'. how i can fix it?
You may need to install Pillow. 'pip install Pillow' should do the work 💪💪
Hi , is that possible to add multiple boundary /square frame thingy to detect multiple object ?
Hey, yes it is possible, you could try by getting the bounding boxes of all connected components, the same way I did in the parking spot detector and counter video. 💪🙌
Hello, I'm trying to create a program that detects and counts objects by color, but now I can only count four colors, these are red, blue, green and yellow. I wonder if I can add the whole color scale to the program, how can I do this? I will be very happy if you give back. have a nice day😊
Hey, I think adding the entire color scale to the program will be very challenging. I would suggest to add more colors incrementally until performance is affected. 🙌
Even small pixels of yellow are detected, from the background, how do I increase the priority to be for just the object
You could apply a blur to remove the small yellow pixels before applying the color detection.
@@ComputerVisionEngineer I am trying, but I cannot find a way. Can you make a small guide on it? I have been trying.
i have a a problem sir when i try to detect a red color for example [0,0,255] it does not work ? i hope u reply and thanks
Hey, maybe you need to try with different forms of red, as specified here www.rapidtables.com/web/color/red-color.html. Have you tried other colors?
amazing video !
Thank you! 😃
like it, really cool video.. the kicker for me was the "BRG" .. it really confused me for a while.. was experimenting with the values there.. was trying to read the color of a led which i thought was green but was more white:ish with a green halo haha, is it possible to detect more colors at the same time ?
Yeah, I hear you, something is tricky to specify the exact hue of a given color. About detecting more than one color at the same time: yes, absolutely! The code can be adapted to detect as many colors as you wish! 😃💪
i want to set up color for red. how ca i do it. i need little help
Just set the color to red lol
Just look up on the internet what's the BGR Colorspace of red
Great video but I have a slight issue it says TypeError: VideoCapture() takes no arguments. I'm assuming this means its having trouble finding my webcam footage but I'm not sure how to fix it
also the 2 in VideoCapture(2) shows up as an unexpected argument
what opencv version are you using? what is your Python version and your OS? are you using an IDE?
@@ComputerVisionEngineer im using Pycharm vers 3.11 on a windows. not sure what version of opencv it is i think its 22.3.1 if that sounds right.
sorry no 4.6.0
Please try to execute it without pycharm, also try with other Python version and other version of opencv.
you are superman
instead of that i want the code to analyse the colour of my face, can you please guide??
when I tried red instead of yellow it didn't detect it even if I try different forms of red. I would really appreciate it if you help me
Hey, I took a look and there was a bug which affected the performance in the case of red hues, thank you for noticing it. It is fixed now, take a look at the repo. 💪🙌
@@ComputerVisionEngineer It worked! thank you so much
What is util here, it isn't available for me?
Can I know what is this software using coding ?
Hello Pasindu, I am not sure if I understand your question, what do you mean? In this video I show you how to detect colors using Python and OpenCV using the HSV colorspace.
pycharm
Great Video! Could you tell me how the code performs
Hey Johannes, I would say the color detection is performing very well! 😃💪
@@ComputerVisionEngineer do you think I could run it on a raspberry pi zero 😂
As long as you can run Python, install OpenCV and get images from a webcam it 'should' work. If you try it on a raspberry pi let me know how it works! 💪
@@ComputerVisionEngineer yes it would be possible to run it but I would need at least 30fps which I don’t think is realistic on a weaker pi.
awesome video man
hey i tried to follow along with your tutorial but for some reason when i try to run it it says IndexError: index 10 is out of bounds for axis 0 with size 1 how can i fix this ?
Thank you
very useful. How can i use this code to detect red color object
Hey, define a new variable called red = [0, 0, 255] and use it instead of the 'yellow' variable. You may need to adjust the definition of red in case is not the exact hue of red you are looking for! 😃🙌
@@ComputerVisionEngineer i tried with a new variable. But hue value is the problem I think. Thanks for the reply
@@microcontroller_garage5387 Hey, I have tried using new variable to detect the red color but there is a problem. Did you managed to solve it?
how should i make it recognise the color red instead of yellow
Hey Jerome, try defining a new variable called red = [0, 0, 255], and use that one instead of 'yellow'. If it doesn't detect the exact hue of red you are looking for you would have to see what are the bgr coordinates of the exact color you want to detect 💪🙌
@@ComputerVisionEngineer Hey, I have tried using new variable to detect the red color but there is a problem. Did you managed to solve it?
@@weaamajoor7420 what problem do you have when trying to detect the red color?
@@ComputerVisionEngineer it stop tracking anything.
Plz make a video on how to fix the media pipe installation error: "No matching distribution found for media pipe"
Hey Shamir, please give me more details about your error. What Python version are you using? What is your OS?
did someone know the RGB value for grey things (type road, circuit, ...)
Do you know which array value I should enter for red?
did you find an answer?
awesome
Glad you enjoyed it! 😄🙌
Very useful tutorial, thanks!
I would like to know if you could help me with a project. Do you offer paid consulting?
The concept of the project is to create a system for selecting and separating stones by shade of green using a Raspberry Pi with a camera. The system captures real-time images through the camera and processes these images to detect the green tones in the stones. Based on the shades of green detected, the stones are classified into different groups (1, 2, 3, 4) according to the defined shade scale.
The image generated by the camera is displayed on a monitor, and each stone is marked with a square that indicates which group it belongs to.
Hey, I may create a paid consultation product soon. Do you need assistance with this project?
how to download util , sir
Hi, i need to make a robot that can avoid green and red obstacles. when the robot detect the green obstacle it should go left and when it sees red obstacle it should go right. and I am kinda lost how to make the camera program, can you help me?🥲
Hey, are you working with Arduino? If you are, search for videos on 'arduino color sensor project', there are plenty of videos on youtube which cover the foundations of how to do it. 🙌
Hi.
Free where i can get utul packages??
Hey, what do you mean with 'utul packages'?
How do you put a circle dot in the middle of the lemon/color? I'm trying to create a posture and gait analysis thesis with colored markers and get measurements (angles/distance), I want to be able to detect it and put a point in the middle of each markers and connect each point with lines. Hoping for your response, thanks in advance!
Hey, about the circle take a look at cv2.circle function, you can compute the coordinates of the center of the color by inspecting the bounding box coordinates. 🙌
Hey! Not sure if you still need the code but here it is anyway:
if bbox is not None:
x1, y1, x2, y2 = bbox
cv2.rectangle(frame, (x1, y1), (x2, y2), [0, 255, 0], 5)
center_x = (x1 + x2) // 2
center_y = (y1 + y2) // 2
frame = cv2.circle(frame, (center_x, center_y), 0, (0, 0, 255), 5) # change the 5 to adjust thickness
just replace the if statement where you draw the rectangle and there should now me a circle in the middle of the object
OK. I found the link
太牛了...附送源代码....老哥你好...能做一个指定区域里的颜色识别监测不
Nice video man
DO I NEED TO INSTALL VSCODE
yes u can use vs code even this ide the one he using
Cezar?????
thankss
😄🙌
Hey, is it possible to make it so the code runs the tracking on screen? Im trying to make an extension that runs on my pc on top of all my apps to track certain colors for premiere keyframes. Im not that skilled in coding so was hopping for a little help.
Hey Sergey, you want to track colors in your screen? Sure, it is possible, I guess the code in this project is a good baseline for that. Take a look at PyAutoGUI library, you can use it to take screenshots, for example. 💪💪
9:25
Is there a way to have multiple bounding boxes when there is more than 1 object of the specified color. Currently is makes 1 large bounding box, which inevitably covers more than the expected color?
How can i take pixels of other colors