From the course: Introduction to Deep Learning with OpenCV

Unlock the full course today

Join today to access over 24,800 courses taught by industry experts.

YOLOv3 in action

YOLOv3 in action

- [Instructor] We can pass either a video or an image through the network. The outputs are stored in the image folder. Now, if I want to pass an image, so we type python yolo.py --image and then the path to the image file. So that's ..images/fruit.jpg And if we head over to the images folder and double-click on fruit YOLO output, we can see the output image from the YOLO version 3 algorithm. Now this is interesting because this is an object-detection algorithm and it has classified the lemon at the bottom-right as an orange and the orange behind the lemon doesn't seem to have been detected at all. Perhaps there isn't enough of the orange visible for it to detect it. Let's see if we can lower the confidence threshold so that it can detect the orange behind the lemon. So I head back to the YOLO code, and I'm going to change the confidence threshold from 0.5 to 0.4. Save the file. Close the image file. Python yolo.py and run the command again. Now if we head back to the images folder, we…

Contents