Ship Detection using DL

[Kshah002]

Deep Learning Simplified Repository (Proposing new issue)

:red_circle: Project Title : Ship Detection from Aerial Images :red_circle: Aim : To detect whether there are any ships in the given picture or not. :red_circle: Dataset : https://www.kaggle.com/datasets/andrewmvd/ship-detection :red_circle: Approach :

Dataset being very scarce the first task to apply data augmentation

Using Transfer Learning approach by leveraging pretrained models - DenseNet121, ResNet50, Xception and EfficientNet B5 and comparing the results. Will try to implement all four models however can assure to use atleast 3 models mentioned

---

📍 Follow the Guidelines to Contribute in the Project :

• You need to create a separate folder named as the Project Title.
• Inside that folder, there will be four main components.
  • Images - To store the required images.
  • Dataset - To store the dataset or, information/source about the dataset.
  • Model - To store the machine learning model you've created using the dataset.
  • requirements.txt - This file will contain the required packages/libraries to run the project in other machines.
• Inside the Model folder, the README.md file must be filled up properly, with proper visualizations and conclusions.

---

:red_circle::yellow_circle: Points to Note :

• The issues will be assigned on a first come first serve basis, 1 Issue == 1 PR.
• "Issue Title" and "PR Title should be the same. Include issue number along with it.
• Follow Contributing Guidelines & Code of Conduct before start Contributing.

---

:white_check_mark: To be Mentioned while taking the issue :

• Full name :
• GitHub Profile Link :
• Email ID :
• Participant ID (if applicable):
• Approach for this Project :
• What is your participant role? (Mention the Open Source program)

---

Happy Contributing 🚀

All the best. Enjoy your open source journey ahead. 😎

[github-actions[bot]]

Thank you for creating this issue! We'll look into it as soon as possible. Your contributions are highly appreciated! 😊

[abhisheks008]

This project is already present in this repo, https://github.com/abhisheks008/DL-Simplified/tree/main/Ship%20Detection%20from%20Aerial%20Images

[Kshah002]

ya i was thinking to work on the same statement and try to get better results with different models as they have used yolo over there. If it is okay ?

[abhisheks008]

Then it's an enhancement of the existing project. You can add your contribution in the existing project folder, no need to create a separate project folder.

Can you share the models you are planning to implement for the enhancement of the existing project? @Kshah002

[Kshah002]

yeah okay so if you could provide me with the link of the folder and also label with necessary gssoc tags, it would be great Here are the models I am planning to implement - DenseNet - https://keras.io/api/applications/densenet/ Xception - https://keras.io/api/applications/xception/ EfficientNet - https://keras.io/api/applications/efficientnet/ ResNet - https://keras.io/api/applications/resnet/

[abhisheks008]

https://github.com/abhisheks008/DL-Simplified/tree/main/Ship%20Detection%20from%20Aerial%20Images

Enhance the model only. No need to remove the files, edit the jupyter notebook and push your codes. Once done update the README file along with your models.

Assigned @Kshah002

[Kshah002]

👍Thank you.

[adityasingh-0803]

i wanna work on this

[abhisheks008]

Please share your approach as per the issue template along with the required details. @adityasingh-0803

[adityasingh-0803]

Data Augmentation: Since the dataset is limited, apply data augmentation techniques like rotation, flipping, zooming, and shifting to artificially increase the diversity of training data.

Model Training: Fine-tune each of the pretrained models on the augmented dataset. The models will be trained to classify whether a ship is present or not in the aerial images.

Model Evaluation: Compare the performance of the models based on accuracy, precision, recall, and F1-score to determine which model performs best for ship detection.

Model Selection: Choose the best-performing model based on evaluation metrics and further optimize it if necessary.

Model Deployment: Save the trained models for later use and further analysis.

[adityasingh-0803]

please tell

[abhisheks008]

Hi @adityasingh-0803 sorry for replying late. I'd like to know about the models you are planning to implement here for this project/issue. You need to implement at least 3-4 Deep Learning models for this project. Also please mention the open source event you are participating in.

[adityasingh-0803]

ok i will work on this

[abhisheks008]

ok i will work on this

Before starting the issue related work, can you please mention the details highlighted in the previous comment?