A Convolutional Neural Network to Perform Object Detection and Identification in Visual Large-Scale Data.

Area of Science:

• Computer Vision
• Artificial Intelligence
• Machine Learning

Background:

• Big data presents significant challenges in analysis, requiring high speed and precision.
• Image data is particularly complex, demanding substantial computational resources.
• Accurate object detection and identification within images are critical computer vision tasks.

Purpose of the Study:

• To develop a deep learning-based method for efficient and precise big image data analysis.
• To address the challenge of object detection and identification in complex image datasets.
• To improve upon the performance of existing state-of-the-art object detection models.

Main Methods:

• A deep convolutional neural network (CNN) approach was proposed for image analysis.
• The CNN model was trained and tested on the Pascal_voc 2007 dataset for object detection.
• Performance was evaluated based on mean average precision and processing speed (FPS).

Main Results:

• The proposed CNN approach achieved 77% mean average precision on the Pascal_voc 2007 test dataset.
• The method demonstrated a processing speed of 16.54 FPS on an Nvidia Geforce GTX 960 GPGPU.
• The approach outperformed established models like R-CNN, Fast R-CNN, Faster R-CNN, and YOLO.

Conclusions:

• Deep convolutional neural networks offer a powerful solution for big image data challenges.
• The developed CNN method provides superior precision and speed for object detection and identification.
• This research advances the capabilities of deep learning in computer vision applications.