Deep-Learning---Computer-Vision

ALL THE PROJECTS ARE IN THE FOLDER NAMED Deep Learning - Computer Vision Using CNN & OpenCV. SOME OF THE CRUCIAL PROJECTS THAT I AM GOING TO MENTIONED HERE. REST OF THE OTHERS FIND IN THAT CERTAIN FOLDER. HAVE A LOOK.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/tree/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV

                                                  Atikul Islam Sajib

---

This is the domain of Computer Vision where the projects built with CNN architecture, RCNN, Fast RCNN, Faster RCNN, Yolo with its version, and OpenCV.

1. Latest Project was Polyp Detectionwith 3000 datasets that downloaded from Roboflow. This was trained with Yolo version 5, custom training. The mPA score is above 89%.

The link of this project is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Polyp_Detection_using_Yolo.ipynb

The video of this project is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/polyp_video%20(1).mp4

---

2. Latest Project was Liver Disease Detection with 4000 datasets that downloaded from Roboflow. This was trained with Yolo version 5, custom training. The mPA score is above 83%.

The video of this project is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Object_Detection_of_Liver_Infection_using_Large_Yolo_model.ipynb

---

3. Warehouse safety measuremt object detection. This dataset was downloaded from Roboflow and the performance is good. The approach that I used is Yolo large version 5 and the mPA is good enough.

The link of this project is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Warehouse_Safety_using_Yolo5_object_Detection_(1).ipynb

---

4. Face mask Detection using CNN architecture and OpenCV. The accuracy was above 98% with recall, precisiona and f1 score is 98, 97, 99% respectively.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Face%20Mask.ipynb

---

5. Leaf Disease Detection using Detectron2 with the model Zoo - Custom Training. I also used Yolo version 5 for comparision the model performance.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Detecting_diseases_Object_Detection_Image_Dataset_using_Yolo_Version_5_Custom_Training.ipynb

---

6. This is the Simple object Detection peoject that is used in TFOD2.x without custom training.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/TFOD2_x_Setup_object_detection_using_pretrained_ZOO_model.ipynb

---

7. This is the project named X-ray classification for the disease named Pneumenia. The approach that I used in CNN architecture, for instance, own architecture, Pre trained model, for example, VGG16, ResNet, InceptionNet. In this pretrained model, the approach that I used is Feature Extraction and Fine Tuning. Then I used the OpenCV model to do the object Detection.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/X-ray%20classification%20-%20CNN.ipynb

---

8. This is the project named Smoke classification. The approach that I used in CNN architecture, for instance, own architecture, Pre trained model, for example, VGG16, ResNet, InceptionNet. In this pretrained model, the approach that I used is Feature Extraction and Fine Tuning. Then I used the OpenCV model to do the object Detection.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Smoking%20and%20Not%20Smoking%20Object%20Detection.ipynb

---

9. This is the project named COVID-19 classification. The approach that I used in CNN architecture, for instance, own architecture, Pre trained model, for example, VGG16, ResNet, InceptionNet. In this pretrained model, the approach that I used is Feature Extraction and Fine Tuning as well using SIGMOD activation function in the last layer. Then I used the OpenCV model to do the object Detection.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Covid%20-%2019%20prediction%20using%20CNN.ipynb

---

10. BREAST CancerClassification with the Image Dataset. Using Normal CNN architecture. The accuracy of this model surpassed the fine tuning model - pretrained, for example, VGG16, ResNet, InceptionNet, MobileNet.

The link is given below:

https://github.com/atikul-islam-sajib/Deep-Learning---Computer-Vision/blob/main/Deep%20Learning%20-%20Computer%20Vision%20Using%20CNN%20%26%20OpenCV/Breast%20Cancer.ipynb

---

PLEASE CHECK THAT FOLDER, YOU WILL GET OTHER PROJECTS THAT I WAS NOT MENTIONED IN HERE.