International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
Real Time Object Detection and Information Extracting System Using Machine Learning Omkar Garudkar1, Sujit Walhekar 2, Aditya Pawar3, Shivansh Sambre4 1,2,3,4 B.Tech in Computer Engineering, Dept. of B.tech Computer Engineering, Ajeenkya D Y Patil University
(ADYPU), Pune, Maharashtra, India. ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In the generation of copious information
such as content extraction, image recognition, and information retrieval from dynamic web pages.
available on the internet, extracting pertinent data from web content has become a critical task for diverse applications. This document presents an innovative approach to object detection via a webcam and Gemini AI, aiming to augment the efficiency of data extraction processes. The proposed system harnesses computer vision techniques to identify and locate objects of interest within web pages, thereby streamlining the extraction of relevant information.
Key features of the proposed system include real-time object detection, adaptability to diverse Gemini AI structure, and compatibility with different objects. By employing machine learning algorithms, the system continuously enhances its detection accuracy through learning from user interactions and feedback. Additionally, the system prioritizes privacy and security, ensuring the utmost care in handling sensitive information during the extraction process.
Key words: Python, Open CV, YOLO, Haar Cascade, Django, Tensor flow, Tessract, Machine Learning, Html, CSS.
The primary objective of this system is to provide a versatile and user-friendly solution for individuals seeking to efficiently extract pertinent data from the expansive realm of the internet. Whether employed for market research, competitive analysis, or content creation, the Webcam-Based Object Detection System has the potential to revolutionize the manner in which users interact with and derive valuable insights from web content. As we delve into that this system represents a substantial advancement in the field of efficient data extraction from the internet.
1. INTRODUCTION: The advent of the internet has revolutionized how we access and utilize information. Consequently, extracting data from web-browser has become a critical task for various applications. As the volume and diversity of online information continue to grow, the demand for efficient and accurate data extraction methods becomes increasingly imperative. In response to this need, we present an innovative solution: a Webcam Based Object Detection System using Yolo and haar cascade algorithm designed to optimize and enhance the process of extracting relevant data directly from Gemini AI.
2. LITERATURE REVIEW: Numerous systems have been developed by engineering students for object detection system using webcam. While these systems are functional, there is a need to improve their efficiency.
Traditional web scraping and data extraction approaches often face challenges due to dynamic web page structures, diverse content formats, and the need for continuous adaptation to evolving websites. Our proposed system effectively addresses these challenges by integrating advanced object detection techniques within the familiar environment. This approach not only simplifies the data extraction process but also unlocks new opportunities for automation and customization.
The existing systems are: i) In existing system, they used a RPN and Fast R-CNN, where high-quality region proposals are generated by end-to-end training of the RPN. And these proposals are then utilized by fast R-CNN for detection. In this system, there are some drawbacks like Detection accuracy remains challenged by factors such as varying lighting conditions, occlusions, and dataset biases. [1]
The system seamlessly integrates with popular yolo3 and haar cascade algorithm, providing a user-friendly interface for individuals and organizations to interact with. Utilizing state-of-the-art object detection models, the system accurately identifies various objects, such as images, and multimedia elements, embedded within web content. This capability is particularly valuable for automating tasks
© 2024, IRJET
|
Impact Factor value: 8.226
ii) In this existing system, they develop a project real-life object detection system using basic webcam. The main motive of the project is it will used in various domains like defense, production line and monitoring In this system,
|
ISO 9001:2008 Certified Journal
|
Page 2525