Видео

Hi @@orientasiprodikelasA, thank you so much! I really appreciate your support and enthusiasm. Glad you enjoyed the video!

@AdiTOSH yeah, I enjoyed your video. Maybe next video for controlling autonomous car using Model Predictive Control?

I’m so glad to hear you enjoyed the video-thank you for your kind words! 😊 Model Predictive Control sounds like a fascinating topic, and I’ll definitely consider it, but no promises just yet. That said, I’m really looking forward to hearing about your progress on the project when you get to it! It’s always exciting to see how these ideas come to life in different applications. Best of luck, and keep me updated!

Hi @@orientasiprodikelasA, thank you for your kind words! 😊 Great question-let me clarify this for you. In this video, the lane detection uses a histogram-based approach and sliding windows to identify lane pixels. Here's how it works: 1️⃣ Histogram: It helps locate the base points of the lanes by analyzing pixel intensity in the lower part of the image. 2️⃣ Sliding Windows: Starting from the base points, sliding windows are used to follow the lane lines vertically and gather lane pixels efficiently. This method is different from the Hough Transform, which detects lines by finding intersections in a parameter space. The sliding window approach is more specific to curved or non-linear lanes, making it better suited for many real-world scenarios like highways. If you'd like, I can make a more detailed comparison of these methods in a future video. Let me know if that would be helpful! 😊

@@AdiTOSH A very good explanation. I want to use this lane detection as a reference path for my control. What meaningful parameters should I take? In Hough Transform, it provides x and y coordinates, but how about this sliding window method? Great video, I’ve subscribed to your channel.

Thank you so much @@orientasiprodikelasA for subscribing! 😊 I'm glad you found the explanation helpful. For using this lane detection as a reference path for control, here are the meaningful parameters you can extract: 1️⃣ Lane Pixel Coordinates (x, y): Just like Hough Transform, you can gather the x and y coordinates of the detected lane pixels from the sliding window method. These can help you define a trajectory or centerline. 2️⃣ Lane Polynomial Coefficients: After detecting the lane pixels, this method fits a polynomial curve (usually a second-order polynomial) to approximate the lane’s shape. The coefficients of this polynomial (a, b, c) are useful for describing the lane curvature and slope. 3️⃣ Lane Offset: Calculate the car's offset from the lane center (typically based on the midpoint between the detected left and right lanes). This is critical for control systems. If you're planning to implement a control algorithm, the polynomial coefficients and offset values are particularly helpful for tasks like path following and steering correction. Let me know if you'd like more guidance, and I'd be happy to assist. Best of luck with your project-it sounds exciting! 😊

@@AdiTOSH yes, I'm interested with the hough transform too for detecting lane.

@@orientasiprodikelasA that's great to hear! 😊 The Hough Transform is indeed a powerful technique for lane detection, especially for straight or slightly curved lanes. I can create a dedicated video explaining how to use the Hough Transform for lane detection, comparing it with the sliding window method, and discussing its pros and cons. Let me know if that would be helpful! Thanks again for your enthusiasm and support-it means a lot! 🚗✨

Thank you so much for your kind words! 😊 I'm glad the video helped clarify the difference between CSP and CORS-it’s definitely a tricky topic, and I can relate to running into issues when they’re not configured properly. It's great that you're diving deeper into this topic with Código Heroe! If you have any questions or if there’s another web security concept you'd like me to cover in a video, feel free to let me know. 🙌🏻 I used a translator app to write the following in Spanish. Please forgive any errors! ¡Muchas gracias por tus amables palabras! 😊 Me alegra saber que el video ayudó a aclarar la diferencia entre CSP y CORS. Es un tema complicado, y puedo relacionarme con los problemas que surgen cuando no se configuran correctamente 😅. ¡Qué genial que estés profundizando en este tema con Código Heroe! Si tienes alguna pregunta o hay otro concepto de seguridad web que te gustaría que cubra en un video, no dudes en decírmelo. 🙌🏻 Usé una aplicación de traducción para escribir esto en español. ¡Perdón por cualquier error!

@@adfinemrising Hi there! Thanks for your question-your OpenCV + Flask project sounds exciting! Deploying it with Docker is a smart idea, as it ensures all your dependencies (like OpenCV) work seamlessly on the server. Let me help you out with this! 1. Docker for OpenCV + Flask: Docker is perfect for packaging your app along with OpenCV dependencies. I recommend starting with a simple Dockerfile to create a lightweight container. In fact, in my video on deploying a .NET web app with Docker on Azure, I show how to set up Docker and use it for deployments. The process will be quite similar for your Flask app: 2. Is Azure Free? Yes! Azure offers a free tier with services like Azure App Service and Azure Container Instances. You can easily deploy your containerized app without incurring costs as long as you stay within the free limits. 3. Is Azure Easy to Use? Definitely! Once you understand Docker (which I can help with), deploying on Azure is straightforward. My video walks through using Docker and Azure together step by step. If you follow a similar approach, you'll have your OpenCV backend running in no time. If you're interested, I can create a tutorial showing: ->How to create a Docker container for your OpenCV + Flask app. ->How to deploy it on Azure using Azure App Service or Azure Container Instances. Let me know if this would help, or feel free to share any specific questions-I’d be happy to assist! 😊

@@AdiTOSH ->How to create a Docker container for your OpenCV + Flask app. ->How to deploy it on Azure using Azure App Service or Azure Container Instances. absolutely!! both would help, thank you very much in advance sir!!

Hi again @@adfinemrising 😊 Thank you for confirming-I'm thrilled that you found my suggestions helpful. I’ll start working on a tutorial covering both topics: 1⃣ How to create a Docker container for your OpenCV + Flask app. 2⃣ How to deploy it on Azure using Azure App Service or Azure Container Instances. In the meantime, if you’d like to get a head start, I recommend checking out my video on deploying a .NET web app with Docker on Azure-it introduces Docker and Azure fundamentals that will also apply to your project: 🔗 ruclips.net/video/akyE98yY-E8/видео.htmlsi=PNuLiHtssmhI6iLJ | Deploy .NET Web App with Docker on Azure in 15 Minutes | Full Guide with DevOps Pipeline Stay tuned, and I’ll notify you once the new tutorial is live. If you have any specific challenges you’re facing right now with Docker or Azure, feel free to share them-I’d be happy to assist further! 😊

@@AdiTOSH i've already subscribed hehe, there not many people who do opencv. so seeing someone doing deployment is a god send! i'll watch this video too so i can familiarize myself with azure and docker. thanks again!!

@@mahdiAsh-k6b Thank you for your comment! If you're asking about detecting and recognizing traffic signs, that's definitely an exciting topic in computer vision. It involves techniques like object detection and image classification using tools like OpenCV or deep learning frameworks. Let me know if you're interested, and I could consider creating a video on how to detect and classify traffic signs. 😊

@@ArmanHudda-w9m For the US visa interview in Kolkata, mobile phones and other electronic devices are generally not allowed inside the consulate premises. However, you can keep your mobile phone in nearby locker facilities or safe deposit services available around the consulate. As per my experience, it is best to leave your phone in hotel room and not bring it if you are going solo. If the venue is the new Pataka House, some visitors have shared that there are local shops or kiosks nearby offering locker services for a small fee. Make sure to confirm their reliability before handing over your belongings. I recommend arriving one day before to explore these options and ensure a smooth experience next day. Let me know if you have any more questions! 😊

@powerpat1 Sure! Here's the YAML file used in the video: drive.google.com/file/d/1CGdbrX1Py6TKNBs-6xNAEumYKke-W4FU/view?usp=sharing Feel free to check it out and let me know if you have any questions or need clarification on any part of it! 😊

@nandagopalchalasani6964 Thank you so much for the update! It's incredibly helpful for everyone planning their trip. I'll make sure to highlight this in the video description and a pinned comment so viewers are aware of the new venue at Pataka House. If you have any additional tips or details about the new location, feel free to share. It’ll definitely help others! 😊

@@laudemirferreira3227 Thank you for your support from Brazil! A CNN-based approach for lane detection is a great idea & and since my channel already covers CNN projects-I’ll definitely consider it for a future video. Stay tuned!

You're absolutely right! Perspective transformation can be used to achieve a bird’s eye view, but it’s also versatile enough for other angles too. 😊

Hi @@priyampratimsaikia4932, thanks for watching the video! The error you're seeing usually happens when the **window size (`win_size`)** used to calculate image quality metrics (like SSIM) is larger than the dimensions of your input image. This can occur if your degraded or source images are smaller than **7x7 pixels**. To fix this, you can either: 1. **Ensure that your input images are at least 7x7 pixels** before processing. 2. **Specify a smaller `win_size`** by adding it to the function call where the error occurs, with an odd value that’s smaller than the shortest dimension of your images. For example: # Add win_size parameter in your SSIM calculation (adjust to your image dimensions) ssim_value = ssim(image1, image2, win_size=3) # Ensure 3 or another odd value works for your image size Also, is the case that you using the code as it is from google colab link shared in the description and getting this error? Let me know if that helps or if you have more questions!

Thanks for helping❤

It's working now

@@priyampratimsaikia4932 Glad I could help! Feel free to reach out if you have more questions. Happy learning!

You're very welcome! I'm so glad the video was helpful for you. Best of luck with your visa application-I hope this makes your travel smoother & easier. Safe travels! Also, I would be grateful and it would be really helpful if you could share the differences from your experience after the travel. Your insights could help others as well!

@AdiTOSH Surely, I'll be more then happy to do that.

I need your help

Hello! How can I assist you? Feel free to share your question or concern here.

Thanks for your feedback! This was one of my earlier pieces, and I’ve worked hard to improve since then. I appreciate your understanding.

👍👌 Thanks!

@@dilipviswanath9106 yes its allowed.

​@@sonaliteachergoa2073 You can choose any nearby hotel with good reviews that fits your budget But it’s best to check recent ratings and prices online. Safe travels!

Thank you for your interest in the code! Unfortunately, I can’t share it as it’s a collaborative effort involving a large team. I appreciate your understanding.

@@AdiTOSH so can you tell me atleast to make such a robot and learning platform

Thank you @@pushpanjalijha2779 for understanding, sure, Here’s a quick guide to get you started: Platform Overview: This project uses ROS (Robot Operating System) and Gazebo for simulation. ROS provides a flexible framework for writing robot software, and Gazebo is a powerful tool integrated with ROS for simulating complex robotic environments. Programming and Libraries: The code is in Python, and for image processing, we use OpenCV. If you're new to OpenCV, here's a guide I put together to help you understand the basics: ruclips.net/p/PLCiTDJays9rXh-TycvwVYHLYNg1JZigVS Learning Resources: 1. For ROS: Start with the official ROS documentation, which has extensive tutorials on setting up and working with ROS nodes, topics, and other essential components. - www.ros.org/ 2. For Gazebo: Here’s the official Gazebo documentation, which covers everything from installation to advanced simulation features. - gazebosim.org/home

Glad to hear it helped with your semester project! Best wishes for your studies! :)

I don’t understand why this could be happening. Try removing entire logic and performing a simple video run. If that too gets the error, there’s some issue with the video

Yes it does. But you might not want to do that. For gray images you can threshold on gray-scale itself. *Example code:* import cv2 import numpy as np from matplotlib import pyplot as plt # Read the grayscale image img = cv2.imread('your_grayscale_image.jpg', cv2.IMREAD_GRAYSCALE) # Apply different thresholding types ret, thresh1 = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY) ret, thresh2 = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY_INV) ret, thresh3 = cv2.threshold(img, 127, 255, cv2.THRESH_TRUNC) ret, thresh4 = cv2.threshold(img, 127, 255, cv2.THRESH_TOZERO) ret, thresh5 = cv2.threshold(img, 127, 255, cv2.THRESH_TOZERO_INV) # Display the results titles = ['Original Image', 'BINARY', 'BINARY_INV', 'TRUNC', 'TOZERO', 'TOZERO_INV'] images = [img, thresh1, thresh2, thresh3, thresh4, thresh5] for i in range(6): plt.subplot(2, 3, i + 1) plt.imshow(images[i], 'gray', vmin=0, vmax=255) plt.title(titles[i]) plt.xticks([]), plt.yticks([]) plt.show()

Sure, the link is there in the video description.

Yes of course, anything thats not electronic and bluetooth.

Hi @acrowfliedover, Thank you for engaging with my video! I appreciate your thoughtful question. Let’s discuss why sorting might not be necessary for this problem. While sorting can indeed simplify certain problems, it’s essential to consider the specific context. In some cases, we can achieve the solution without sorting. Here’s how: *Problem Context:* First, let’s understand the problem requirements. If the problem involves counting permutations (rather than combinations), sorting might not be crucial. Combinations focus on selecting items without regard to their order, whereas permutations consider order. *Example:* We can see in the video from 4:30 to 6:30 on impact of sorting. *Efficiency:* Sorting can be computationally expensive, especially for large datasets. If we can avoid it without compromising correctness, that’s a win! *Trade-offs:* However, there are trade-offs. Sorting might simplify the problem conceptually, making it easier to reason about. It could also lead to more efficient algorithms in some cases. In summary, I believe sorting truly affects the problem outcome. If you have further insights or disagree, I’d love to hear them!

Any one help me to run the code broo

Yeah bro, you should be able to directly run the code, just follow along with the video, it should be fine. You will not get any error.

@@AdiTOSH image is not downloading broo

That's the problem

@@My-bself I see, that should not be happening. It does download for me, can you try this link in that case, drive.google.com/drive/folders/1e5u38BG6lBjUm24uRb4WS3lmms0r6UtM?usp=share_link

Yes, absolutely. You can use it for a set of 200 or more given thats your testing set. For training, we need a much larger set and we are using pre-trained weights for the sake of avoiding high gpu consumption that training typically takes. The SRCNN model is a basic model with just 3 neural network layers. In the world of image super resolution, there are way more deep neural networks with a huge number of layer, way more powerful models that result in astonishingly sharp changes on low resolution images. This model is usually used to learn to fundamentals and understand how image super resolution works. The intention is not to use it to actually perform super resolution on real images captured by our cameras with huge number of pixels. But for less number of pixels, on standard test set, the model performs well even on visual levels. Even, google phones, use image super resolution as post processing to produce picture qualities which are impossible to capture without dslr camera like huge lenses. But the models they use surpasses this basic model that we are studying as a part of research by leagues. However, like any technology, the fundamentals stays the same and without them you cannot comprehend the versions built on top of it. Thus, from study point of view, research point of view, this is an important model, but of little use in actual image super-resolution tasks on pictures captured by cameras. I hope this generates clarity over your concerns, thanks!

@@AdiTOSH ooooho ok , now I got it. thank you.

Hello @DinosaurRex-tw8mn, allow me to help you generate clarity on this. We have 8 pairs in total, namely, (x1,y1), (x2,y2), (x3,y3), (x4,y4), <- initial points (x1', y1'), (x2', y2'), (x3', y3'), (x4', y4') , <- transformed points Now, For initial points, these are the points we have chosen, so they are not variables but known coordinates as we ourselves have chosen. These are the coordinates of the trapezoid. It could be anything, like, (x1', y1') = (100, 100) (x2', y2') = (60, 150) (x3', y3') = (180, 150) (x4', y4') = (200, 100) Also, we know the transformed points to be on the corner of the screen and therefore we know them as well, i.e, (x1', y1') = (0, 0) (x2', y2') = (0, 480) (x3', y3') = (640, 480) (x4', y4') = (640, 0) Now, we have the relation from bilinear transformation, x1' = c1.x1 + c2.y1 + c3.x1.y1 + c4 x2' = c2.x2 + c2.y2 + c3.x2.y2 + c4 x3' = c1.x3 + c2.y3 + c3.x3.y3 + c4 x4' = c1.x3 + c2.y3 + c3.x3.y3 + c4 Here only c1, c2, c3 and c4 are unknown as rest are known. So, we have 4 equations and 4 unknowns. We can use these to find the values of the constants c1, c2, c3, c4. The entire idea is to find constant values using set of know coordinates both initial and transformed and then once we have the constant values, use the same four equations to find transformed coordinates for other set of input coordinates for which we do not know the transformed coordinates. I hope this helps you, thank you for giving you time, happy learning. :)

@@AdiTOSH Great Stuff, helped a lot.

Hi @ramakrishnamiryala, The link does not open a webpage but instead downloads the zip directly through your browser. Can you please check if your browser if blocking the download or try a different browser. Thanks.

Thank you @omercandemirci8580, Well the answer is yes, you can get the angle and I believe that it can be used to program an autonomous toy sized to cycle-size robotic car but not an actual one. However, one of the resources that you can use to get the steering angle from the above code: github.com/georgesung/advanced_lane_detection I hope this helps, thanks for watching, happy learning.

Welcome @cc-ut7ow! While perspective transformation changes the view, distance transformation is a different concept. Remember that perspective transformation is a non-linear process, and it’s essential to understand the context and purpose of your application. If you’re interested in measuring distances, consider other techniques like stereo vision or depth estimation

Welcome :)

Hi @Lachu-hi6gg, Sure, here is what I am aware of: EDSR, WDSR, and SRGAN: ◦ The GitHub repository provides implementations of EDSR, WDSR, and SRGAN for single image super-resolution. You can load pre-trained weights and apply super resolution to images 2. ◦ Check out the code and examples in the repository here. github.com/krasserm/super-resolution

Thank you, will look along the suggested lines.

Most welcome! About creating your own h5 file, I am afraid I could not do it either. I observed that google colab is not enough to train SRCNN model and my computer has neither the amount of GPU that training requires. I am afraid I cannot help you here. Thanks.

Yes we can do the same using raspberry pi. But in the approach shared, we did not make use of any ml models, but can we dump ml model to Raspberry pi is a different question. Thanks!

@@AdiTOSH ok thank you

@@bharathvarun3407 Most welcome, happy learning!

Agreed. Great observation! Glad you loved it, most welcome.