Road Image Selection Using Capturing Situation Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies do not effectively improve the accuracy of statistical information about traffic on a road or the recognition of recognition targets in image analysis.
Innovation Solution
An information processing system and method that utilizes a machine learning model to generate capturing situation information from images of roads, incorporating capturing apparatuses, a generation unit to process images, and a storage processing unit to store images based on this information, enhancing the accuracy of analysis results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning models are used to analyze road images, then analysis capability is enhanced, but accuracy of statistical information and target recognition remains insufficient without effective utilization of capturing situation information
Solution Approach 1:
The system performs preliminary classification of captured images based on capturing situation information (time, weather, road type, camera position) before machine learning analysis. This preliminary action filters and organizes images into relevant categories, ensuring that the machine learning model processes only appropriate images for specific analysis tasks, thereby improving both recognition accuracy and statistical reliability.
Solution Approach 2:
The system changes the parameters of image selection by incorporating multiple capturing situation parameters (capturing timing, capturing condition, apparatus information, road information) into the filtering process. This multi-parameter approach transforms the image selection process from simple random or sequential selection to a targeted selection based on relevant capturing conditions, enhancing the quality of training data and analysis results.
2Quantity of substance
If all captured images are stored and analyzed, then comprehensive statistical information is obtained, but processing time and computational resources increase significantly
Solution Approach 1:
The system extracts only the necessary capturing situation information (time, weather, road type, camera position) from each captured image metadata. By extracting and utilizing these key parameters, the system filters the large volume of captured images to identify only those relevant for specific analysis tasks, significantly reducing the number of images that need to be processed while maintaining comprehensive statistical accuracy.
Solution Approach 2:
The system segments the image processing task by first classifying images based on capturing situation parameters, then processing each segment with appropriate machine learning models. This segmentation approach divides the large-scale processing task into smaller, manageable segments that can be handled more efficiently, reducing overall processing time while maintaining comprehensive analysis coverage.
3Ease of manufacture
If machine learning models are trained with unfiltered captured images, then model training is simplified, but accuracy of analysis results deteriorates due to irrelevant or inappropriate images
Solution Approach 1:
The system performs preliminary filtering of captured images based on capturing situation information before feeding them to machine learning models. This preliminary action automatically selects images that match the training requirements (e.g., selecting only nighttime images for nighttime detection models, or only images from specific road types for road condition analysis), ensuring high training data quality without requiring manual image selection while maintaining training simplicity.
Data Source
AI summary
To improve accuracy of a result acquired by analyzing, by using a machine learning model, an image in which a road is captured, an information processing system 100 is an information processing system for collecting a machine learning image, and includes a generation unit 122 and a storage processing unit 123. The generation unit 122 processes an image in which a road is captured, and generates capturing situation information indicating a situation related to capturing of the image. The storage processing unit 123 performs processing for storing an image, based on the capturing situation information.