Object Detection OpenCV Python | Easy and Fast (2020)
Вставка
- Опубліковано 29 сер 2020
- In this tutorial, we are going to learn how to detect objects using OpenCV and python. The Object Detection OpenCV method we will use is a sweet balance between speed and accuracy. Unlike many other object detection methods, we will be able to run this in real-time with a good amount of accuracy.
🚀🚀 My Urdu/Hindi AI UA-cam Channel 🚀🚀
/ @murtazahassan01
Download Code:
www.computervision.zone/cours...
Premium Courses:
✔️ Computer Vision Game Development Course:
bit.ly/3ttLZ2s
✔️ Computer Vision with Arduino Course:
bit.ly/3wzLB4m
✔️ Advanced Drone Programming Course:
bit.ly/3qs3v5g
✔️ Learn to Build Computer Vision Mobile Apps:
bit.ly/3uioY1J
✔️ Jetson Nano Premium Course:
bit.ly/3L8uIlF
⚙️⚙️⚙️--My Gear - ⚙️⚙️⚙️
👉 Complete Gear 💈: www.computervision.zone/tech-...
👉 My PC Specs 🖥️: www.computervision.zone/tech-...
👉 My Video Shooting Gear📽️ : www.computervision.zone/tech-...
👉 My Laptops 💻: www.computervision.zone/tech-...
👉 Educational Products🧑🎓: www.computervision.zone/tech-...
👉 UA-cam Starter Kit 🔴: www.computervision.zone/tech-...
Follow Me:
TikTok: bit.ly/3Vo76OQ
Facebook Group: bit.ly/3irDcb7
Discord: bit.ly/3JvyxAM
Facebook Page: bit.ly/3IvpU7W
Instagram : bit.ly/3NdGME3
Website: bit.ly/3ICFTS0
Github: bit.ly/3woU6PS
#ComputerVision
#OpenCV
#CVZone
This was great, so easy and quick. I watched so many other tutorials and there are always a million dependencies and errors, this was awesome, thanks!
how to download COCO dataset?
Peter, could you possibly help me with fixing my error?
i am having issues in getting the resource files such as ssd mobile net v3 and frozen inference....can u plz help me
This was super helpful! Your live coding is very impressive. Thanks!
Wonderful tutorial. Thank you so much for sharing your knowledge for free, you make the world a better place.
You are so welcome!
Good Job Murtaza. Hello from Turkiye to Pakistan
8:48 "Sometime time it good to be lazy" quote of the day😁
I went on your website to see the weight files and couldn't find them. Where are they?
yes...I also couldn't find it
www.computervision.zone/topic/basic-code/
Those are paid 😥
Thanks so much for making this, I learn't a lot from you.
Thank you very Much. This has been very very helpful!
Awesome Sir ...... This the one of the best video i seen in youtube......you code are esay to understand.....thank so much sir... That One is really esay to use .....
thank you soo muchh!Very informative and detailed video
I will definitely put this to use in my own project sometime! Thank you for making this tutorial.
Can i get your project name?
This is just gold. Thank you so much Murtaza I'm a broke student but eager to dive into Computer Vision and thanks to your altruism in making this content I am can to learn something that otherwise I would not be able to. Thank you
i am having issues in getting the resource files such as ssd mobile net v3 and frozen inference....can u plz help me
great tutorial from a great teacher
Thanks for the video! Love it! Can you do a tutorial on adding custom classes to the coco dataset for Object Detection with OpenCV Python? :)
Incredible... Thanks for the effort and time
Great Tutorial Murtaza .from Iran to Pakistan
need weekly thrice videos like these please I am sharing this with all my friends
You sir, are a god amongst men.
Excellent explanation. You made me to fell in love with openCV.
what means box[2] and box[3] ?
Thanks Really Helpful
I love 🌹 your videos
Interesting cant wait!
Great video!! everything worked perfectly
can u pls explain how to do it
Excellet tutorial! Can you please explain why we have set the input parameters(Size, Scale, Mean, SwapRB) to the net object?
i am having issues in getting the resource files such as ssd mobile net v3 and frozen inference....can u plz help me
Merciiiiii pour votre effort
you are BEST . tanx for sharing .
If you would like to follow along, here are the prerequisites:
1. Pycharm IDE with opencv installed
2. Coco.names file
3. Config and Weights file
4. Test image or webcam
All the files can be downloaded from the below link under the "Intermediate" section by the heading "Object Detection Opencv"
www.murtazahassan.com/courses/opencv-projects
" classIds, confs, bbox = net.detect(img,confThreshold=thres" this code getting error" what can i do
there is nothing inside "-----Intermediate----" section??
I can't find the files under the intermediate section. Can u please share them again? We love the video, it rocks!
The weights Are in the dataset folder downloaded from kaggle
classIds, confs, bbox = net.detect(img, confThreshold=thres)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
i am getting error
Nice Doing! Thank you!
Thanks a lot for this great tutorial. I have watched couple of the videos here in UA-cam, but no one make it straight forward like you. In line 22 why you didn't flatten the bbox when you use it with the zip function?
That great. very nice. I have one question how we can compute its confusion matrix, precision, recall with intersection over union IoU in Python?
Thanks!
Thanks for your great work! Your videos never disappoint...
awesome thanks for the tutorial
You are a hero!
For real man you inspired many of my main projects on AI. Thank you
Dr. Murtaza, I need to know how to close up the 'complete' course in your website. At times, I play the video straight from youtube. when I log in in my account it is showing not complete. My question is, do I have to log in to continue the lesson? there is no option to 'continue' later...
Thanks a lot for the tutorial, as usual well explained and straight to the point. Can you please point me to where I can download all the files you have imported at the start of the lesson. I can't find them. Thanks
Did u get them??
@@mohammedkamran8559 yes i need them'
Can you please give straight link to the dataset and the files you add to project at the start of video
Sir can you put the links of the resources files in the description, please?
We cannot perform anything without these files.
Did u get those files??
@@mohammedkamran8559 did you get them?
@@mohammedazzan7529 ys
@@krizz5825 how?
How to connect the camera to the laptop? In order for it to access the program. And does this needs a microcontroller?
Can I do this using a hikvision IP camera rather than webcam? If so then can you do a quick tutorial on this? Thanks for your hard work. It is much appreciated
Thank you so much sir!!!!!!!
Excellent tutorial.! Can you please make a tutorial on how to train Object Detection models using our own dataset?
towardsdatascience.com/creating-your-own-object-detector-ad69dda69c85 u can check here!!
Chekc the ai guy gethub he is explaining everything good but not perfect,
I need a video as you said to
@@shaik.gulshanmohiddin5966 hey shaik, im having some issues with the installation, could you help me solve it?
cannot import name 'model_lib_v2' from 'object_detection"
It worked thanks alot
love your work from bangladesh
Assalamualykm bro yr videos help me alot ♥
Hello, sir a nice tutorial. but can ask if object detection can light LED after the detection is done? I'm trying to apply it in RPI and in real-time
Nice, thanks :)
permisiion to learn sir. thank you
I'm getting error in net = cv2.dnn_DetectionModel....... what should i do?
I gotta say, this video was just what I was searching for, great work man!!!
I have error in classIds,confs,bbox=net.detect(img,confThresholf=thres)
Assertion failed, !ssize.empty() in function
What should I do?
i dont understand why videos like this still get people who dislike it......
Thanks you very much for this video. I hope you will guide custom Data Mobile net ssd
Thank u verry much i love you❤️❤️
Thanks a lot bro.
thank u sm!
Does this work anymore? I'm getting an error for the following.
net = cv2.dnn.DetectionModel(weightsPath, configPath)
AttributeError: module 'cv2.dnn' has no attribute 'DetectionModel'
which function I need to add if I want to count how many times I lost my object from screen
Sir , what if we wanna detect only a spesific object such as pedestrian or car
what kind of methot should we use ?
Thanks for lecture, Did you run Pycharm to using code?
Hi can i know, how to make object identification for trash bin, example, my camera detects platics,the shutter for plastics compartment will open. And vice versa
Good work keep it up bro.
Where can we found the fronzen_interference_graph? Thank you
I am getting this error:
net = cv2.dnn_DetectionModel(weightsPath,configPath)
AttributeError: module 'cv2.cv2' has no attribute 'dnn_DetectionModel'
Do you know or anyone know how to solve this problem? thank you.
is there any way to train new image, or for example i want to classify different types of cakes, please reply sir I really want this for my project
How can i get the width or height of the rectangle in pixel? I want to use it to measure distance along with detetecting the object
I love your videos, maybe do you have a video of cross line detection?
simply awesome
How can I Apply this object detection model on a reception robot and detect the floor to be able to walk and avoid objects ?? Any idea :) Thanks
jazakallahu khoyron bro
Can I use my custom objects for detection like Nuts and Bolts? And can I use that model in R-Pi 4GB?
Thank you! I have a question. How can I get the config and weight files?
I have the same question
Follow the first link in the discription. And enrole it's free.
I love it this video. GREAT JOB :)
Murtaza's you are an expert in this field I need your help How can I custom and training my own SSD MOBILENET ? I want to recognize my own object but with a CNN like this video i don't want to use the Haar Cascade method. Plese I'm working in my last project and for me is more cool using SSD MOBILENET with a predictions instead using HaarCascade. Please could you made a video training your own SSD MOBILE NET please, please, please
SystemError: returned a result with an exception set
Sir can you help me out with this? I've searched for this error in several places but can't seem to be of help...
I'm using VS Code as a platform...
Thank You
How to get the initial files you put in to project? Coco names and SSD files.
I wanna detect some electronic materials like mosfet or transistor or resistor.What should I do for this?Thank you.
Request Tutorial:
- Detect rotation angle from template.
is it possible to include more objects to be detected? like some custom trained objects
As always You're Crack !! @Murtaza's .....
It's work! thanks a lot!!!
Excellent tutorial. I'm currently working on a project that requires an ai to be able to recognize 4 different objects and then determine their orientation orientation this helped a lot with recognition.
Glad it was helpful!
@@murtazasworkshop It indeed was. The big challenge is now how to determine how the object is oriented relative to the camera. Found a lot of papers online but no real explanation on how to do it.
@@murtazasworkshop Where can i download the coco dataset from? And the .pb file? please
I am also working on a project, will this program work on Arduino?please reply
I have error in classIds,confs,bbox=net.detect(img,confThresholf=thres)
Assertion failed, !ssize.empty() in function
What should I do?
Since your videos are always easy to follow and efficient, we are eagerly waiting sir. Thsnk you for working hard to bring excellent tutorials. Can you pls cover a subject how to control relay or any device from pyCharm running on windows? (Bcs of GPIO, I will be able to do in Raspberry)
Hi . There is a similar tutorial i have done that communicates with arduino using serial communication and controls a servo . You can check that. ua-cam.com/video/TymXJVk-I6Q/v-deo.html
@@murtazasworkshop Thank you very much sir.
i am having issues in getting the resource files such as ssd mobile net v3 and frozen inference....can u plz help me
@@parthverma7089 did you get the solution? i too am having problems
classIds, confs, bbox = net.detect(img, confThreshold=thres) i am getting error over here
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^@@murtazasworkshop
Is it possible to extract the deteted box with an object as a separate image? I need it for my project and hoped it could work this way but I am lost....
thank you the video. I wonder where we train the model?
hi its good tutorual but can you tell me how I add anthor coco names like names that were not in coco name if I add that can be destroy the code?
Can you do a video on how to train new objects from a collection of images? Your content is great!
i am having issues in getting the resource files such as ssd mobile net v3 and frozen inference....can u plz help me
What cam are you using?, Are you using a ESP32?
Good Job from India to Pakistan!
(Others are saying you live in pakistan)
Hi, is it possible to use opencv to do custom object detection? Thank you.
Is coco dataset tuple and can i modify it as i need to person to be detected like person one and person two so is that possible ?
Sir after starting first 4 line it shows an error imshow() takes 1 positional arguments but 2 were given ....
Sir plz solve my problem
what is the difference between your method and tensorflow?
My research in Fully convolution Network.. in small object detection
so from. where I should be start.. please help me...
give me some advices
Can you please make a video about scanning the area by LIDAR?
What is the algorithm used (cascade classifier etc.)? How does it detect objects?
Can it be used for a specific object? Thank you, good job.
it can be used for a specific object. For only one object you can create your own haar cascade or for a lot of object you can train your own dataset.
I have error in classIds,confs,bbox=net.detect(img,confThresholf=thres)
Assertion failed, !ssize.empty() in function
What should I do?
SystemError: returned a result with an error set
Can you please help me in this case.
I solved it by downloading same files as Murtaza used (there is a link in the descriptions)
I am not able to open any file from the link in the description, please help
@@pranavsundriyal1251 you need to enroll the course first
@@501ismail501 where
@@501ismail501 where is it man wtf>?????????
Hi! it was such a great tutorial but you haven't attached the .pb and .pbtxt files link, it will be great if you can provide the link.
Help pls. When it detects more than one object in gives out an error: cv2.putText(img, classNames[classId-1].upper(), (box[0]+50, box[1]+30), cv2.FONT_ITALIC, 1,
IndexError: list index out of range
Edited: in fact it shows few objects (with bad accuracy) for a few secs and then gives an error because classIds has index 90 and that is out of coco.names range
same
is there any fix?
I am getting the same error but with numbers 44 and 77... which is definitely within the coco.names range.... for those having issues with over 90 being detected, make an entry in coco.names for that number and see if it fixes it??
where are you getting the coco.names file from? I cant locate your files directly from the link you have provided in the video desciption.
Follow the link in the description and enroll in the course at the top of the page to get access.
This is my first video of its and it's the third time in watching it today so I pressed the big red button directly
i also Get This Problem
SystemError: returned a result with an error set
how to slove it ...??
install package opencv-contrib-python and then it works
@@juanpablomarintrujillo6838 I solved it by downloading same files as Murtaza used (there is a link in the descriptions)