Object recognition
Download as PPTX, PDF1 like164 views
This document summarizes a project on real-time object recognition using Tensorflow. The project uses an Android application and Tensorflow to identify objects in frames from an environment's camera. It trains on datasets of various objects like cars, bikes, flowers and people. The initial results show some misclassifications but accuracy improves after adding more training data. Potential applications of this object detection technology include industrial processes and security systems. The document outlines the project's objectives, methodology, datasets used, results and scope for future enhancements.
Object recognition
- 1. Real-Time Object Recognition Using Tensorflow Presented by :- • Madhavi Soni • Aakanksha Singh • Shreya Patkar Under The Guidance :- Prof. Asmita Deshmukh K.C.College of Engineering & Management Studies & Research.
- 2. What is Object recognition • Object recognition – technology in the field of computer vision for finding and identifying objects in an image or video sequence. Humans recognize a multitude of objects in images with little effort, despite the fact that the image of the objects may vary somewhat in different view points, in many different sizes and scales or even when they are translated or rotated. • Object recognition is a task of finding and identifying object in an image or video sequence. • The goal of instance-level recognition is to match(recognize) a specific object or scene
- 3. • Starting from an image of an object of interest (the query), search through an image dataset to obtain (or retrieve) those images that contain the target object.
- 4. • This application allows the user to learn about the object without needing a text based search. • The main objective of this project it to identify the object in every frame of the environment.
- 5. Existing system • Google goggles
- 7. • The application is build using Android and Tensorflow, an api which has pre-defined Libraries and Machine Learning Algorithms.
- 8. Tensorflow
- 9. We have added a few dataset lf Cars, bikes, flowers, person
- 10. Dataset • Here is the dataset for flowers.
- 11. Dataset • Dataset of Cars.
- 12. Dataset • Dataset of People
- 13. Datasets • Dataset of Bicycle
- 14. • It requires an internet connection. • It requires large dataset.
- 15. Manufacturing Industry Object detection is also used in industrial processes to identify products. Say you want your machine to only detect circular objects. Security In the future we might be able to use object detection to identify anomalies in a scene such as bombs or explosives
- 16. Results • Case 1 : Result when Training • The bottle here is recognized as a Vase.
- 17. Result • Another test case recognizing Laptop as a Tv in Scene 1 • In scene two the same laptop is recognized as a Book.
- 18. Result • Case 2 : Test case scenario after training • After feeding more datasets, we finally get the accurate results.
- 19. Working
- 20. Working
- 21. • At the end of this project, we have developed object recognition application. We have shown that all the objects in the environment are classified into their respective groups. In terms of accuracy, the rate is higher for different kind of objects . It is because the classifier does not get confused when the object are metamorphically distinguish. This system is variant to lighting condition, background scenario and view point. Once the object is detected by the camera using this application, the information relevant to that object is displayed.
- 22. • In the further development of this android app, we plan to make the app more advance by adding language preference as per one's need. • User can save the information the information • User can share the information.
- 23. • Tawfiq A. Al-asadi and Ahmed J. Obaid 2, “ Object detection and recognition by using enhanced Speeded Up Robust Feature”, International Journal of Computer Science and Network Security, VOL.16 No.4, April 2016 • Khushboo Khurana, Reetu Awasthi “Techniques for Object Recognition in Images and Multi- Object Detection” • Nidhi “Image Processing and Object Detection” International Journal of Applied Research ,2015 • Shakuntala Satyawana, Surendra Kumar Agarwal “A Review Paper on Image Segmentation and Object Recognition Procedures” International Journal of Science,Engineering and Technology-2016, Volume 4 Issue 1 • Pankaj Kumar Gautam, Divya Patel “ Object Detection for Improve the Classification Accuracy and Robustness using different Techniques”, International Journal of Computer Applications (0975 – 8887) Volume 112 – No 11, February 2015 • Garrick Orchard, Cedric Meyer, Ralph Etienne-Cummings Fellow, IEEE, Christoph Posch Senior Member, IEEE,Nitish Thakor Fellow, IEEE, Ryad Benosman “HFirst: A Temporal Approach to Object Recognition”, IEEE, Aug-2015