Lane Pattern Detection Using Split-Image Neural Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane detection techniques for autonomous vehicles are complex, expensive, and prone to false detections due to variations in road scenes, and fail to accurately identify lane segments when markings are absent.
Innovation Solution
A method and system that divide an input image into two portions based on lane markings and resize them for independent analysis using a simplified deep learning technique, reducing complexity and computational cost while improving accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If deep learning based lane detection techniques are used, then detection accuracy is improved, but system complexity and computational cost increase
Solution Approach 1:
The patent divides the input image into multiple regions (e.g., left lane region, right lane region) and processes each region independently through separate neural networks. This segmentation approach reduces the complexity of each individual network while maintaining overall detection accuracy, as each network only needs to handle a specific portion of the lane detection task rather than the entire image.
2Ease of manufacture
If heuristics and post-processing techniques are used for lane detection, then implementation simplicity is improved, but detection accuracy and reliability deteriorate
Solution Approach 1:
The patent replaces traditional heuristic rules and post-processing techniques with a deep learning-based neural network approach. The neural network automatically learns lane detection patterns from training data, eliminating the need for manual heuristic programming and post-processing steps, thereby improving both accuracy and reliability while maintaining implementation feasibility through standardized deep learning frameworks.
3Productivity
If existing lane detection techniques are used, then processing can be performed, but false detections increase when lane markings are absent
Solution Approach 1:
The patent performs preliminary processing by dividing the input image into distinct regions before lane detection. By pre-segmenting the image into regions where lane markings are more likely to appear, the system prepares the data in a way that helps the neural network focus on relevant areas, reducing false detections in regions where lane markings are absent while maintaining processing efficiency.
Data Source
AI summary
The present disclosure is related in general to field of machine learning and image processing, that provides a method and system for detecting lane pattern. A lane classification system receives an input image comprising lane markings and co-ordinates of the lane markings, which is divided into two portions, a first image portion and a second image portion based on the lane markings and the co-ordinates of the lane markings. Further, the first image portion and the second image portion are resized into a predefined template size. The lane classification system detects a lane pattern in the two portions by analysing each of the one or more lane markings located in the resized first image portion and the second image portion, independently, based on historical lane data.