Object Detection and Identification via Segmented Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object detection and identification technologies in computer vision struggle to accurately identify objects within images and provide sources for those objects, especially in dynamic environments like advertisements and media, and do not offer seamless acquisition options for users.
Innovation Solution
A system and method that uses a computing device to receive images from user devices, classify them, detect salient objects, identify sources, and facilitate acquisition through a cloud-based platform with modules for image capture, classification, detection, identification, source determination, and acquisition verification, utilizing machine learning algorithms and neural networks for object recognition and saliency detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If object detection and identification technologies are used in dynamic environments like advertisements and media, then object recognition capability is improved, but accuracy in identifying objects and their sources deteriorates
Solution Approach 1:
The system segments the image processing task into distinct modules: image capture, classification, detection, identification, source determination, and acquisition verification. Each module handles a specific aspect of the problem, with the detection module focusing on locating objects and the identification module focusing on recognizing and sourcing them, thereby improving overall accuracy in dynamic environments
Solution Approach 2:
The patent introduces an intermediary classification module that processes images before detection and identification. This classification step acts as a mediator that prepares and filters image data, enabling the subsequent detection and identification modules to operate more accurately on pre-processed and categorized visual information
2Adaptability or versatility
If multiple modules are integrated for image capture, classification, detection, identification, source determination, and acquisition verification, then functionality is improved, but system complexity increases
Solution Approach 1:
The computing device is designed as a universal platform that performs multiple functions through integrated modules: capturing images, classifying them by type, detecting salient objects, identifying objects and their sources, and verifying acquisition. This multi-functional architecture allows a single system to handle the complete object recognition and sourcing workflow without requiring separate dedicated systems for each function
3Ease of operation
If source identification and acquisition facilitation are added to object detection, then user service capability is improved, but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-processing images through classification before detection, and by pre-identifying potential sources during the identification phase. This preliminary processing organizes and prepares data in advance, so that when source identification and acquisition facilitation are needed, the system can quickly retrieve and present relevant information without extensive additional processing time
Data Source
AI summary
An embodiment of the invention may include a method, computer program product and computer system for object detection and identification. The method, computer program product and computer system may include computing device which may receive an image from a user device. The image may be a screenshot captured by the user device from a display. The computing device may classify the image based on features present in the image and detect a salient object contained within the image. The computing device may identify the object in the image and one or more sources of the object in the image.