Wide-Angle Object Recognition Using Far and Nearby Dictionaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The appearance of recognition targets in cylindrically corrected wide-angle images varies significantly between distant and nearby objects, hindering efficient object recognition, particularly in vehicle control systems using wide-angle cameras.
Innovation Solution
A recognition processing device and system that utilizes a far dictionary and a nearby dictionary for each pattern, dynamically recombining them based on the likelihood of recognition, to enhance object recognition efficiency in wide-angle images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cylindrical correction is performed on wide-angle images, then the image distortion is corrected, but the appearance of recognition targets differs significantly between distant and nearby objects, hindering efficient object recognition
Solution Approach 1:
The recognition processing is segmented into two distinct phases: a first recognition phase using far dictionaries for distant objects, and a second recognition phase using nearby dictionaries for nearby objects. This segmentation allows each phase to be optimized for its specific distance range, resolving the contradiction between maintaining recognition accuracy across varying distances and reducing processing complexity.
Solution Approach 2:
The system dynamically switches between far dictionaries and nearby dictionaries based on the detected distance of recognition targets. By making the dictionary selection dynamic rather than static, the system adapts to different distance conditions, maintaining high recognition accuracy while avoiding the complexity of processing all objects with a single universal dictionary.
2Measurement precision
If multiple dictionaries for different patterns are used, then recognition accuracy for various patterns is improved, but processing time increases
Solution Approach 1:
Instead of using all available dictionaries for every object regardless of distance, the system applies partial action by selecting only the appropriate subset of dictionaries (far dictionaries for distant objects, nearby dictionaries for nearby objects). This reduces the excessive action of processing all patterns uniformly, thereby decreasing processing time while maintaining high recognition accuracy through selective dictionary application.
Solution Approach 2:
The system changes the parameter of dictionary selection based on the distance parameter of recognition targets. By adjusting which dictionaries are applied according to the distance parameter, the system optimizes the balance between recognition accuracy and processing time, avoiding the time loss associated with uniformly applying all dictionaries to all objects.
Data Source
AI summary
The present technology relates to a recognition processing device, a recognition processing method, and a recognition processing system capable of realizing more efficient object recognition.The dictionary data storage unit stores a far dictionary for each pattern to be recognized and a nearby dictionary for each pattern, and the recognition processing unit executes recognition processing of recognizing an object included in a wide-angle image captured by a wide-angle camera using the far dictionary and the nearby dictionary. Further, the recognition processing unit recognizes the object by using a plurality of patterns of the far dictionaries and a nearby dictionary of the same pattern as the far dictionary of a high likelihood pattern in which the object is recognized with high likelihood among the far dictionaries. The present technology can be applied to, for example, an image recognition chip that performs pedestrian recognition on a wide-angle image.