Image Sensor Search Control Using Subject Probability Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic shooting cameras face challenges in quickly finding subjects after changes in installation location, as the importance levels of subjects calculated based on spatial coordinates become invalid.
Innovation Solution
An image shooting control apparatus and method that utilizes an image sensor to acquire data, detect subjects, create an environmental map, store history information associating subjects with divided areas, and control the image sensor's orientation to efficiently search and shoot subjects based on probability of existence and historical data, even after changes in installation position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the importance level of the subject is determined based on the installation location of the automatic shooting camera, then the subject search can be optimized for that specific location, but the importance level becomes invalid when the camera position changes
Solution Approach 1:
The space is divided into multiple divided areas, and the subject search is performed by segmenting the overall search space into manageable portions. Each divided area can be independently evaluated and searched based on history information, allowing the system to adapt to different camera positions by re-evaluating the importance level of each segment rather than relying on a fixed installation-location-based importance map
Solution Approach 2:
The system performs preliminary actions by storing history information about subjects in divided areas before the camera position changes. This history information is used to determine the probability of subject existence in each divided area, enabling the system to quickly adapt to new positions without losing previously learned spatial distribution patterns of subjects
2Reliability
If the automatic shooting camera searches the entire space uniformly, then it ensures complete coverage, but the time required to find the subject increases significantly
Solution Approach 1:
The system applies local quality by determining the probability of subject existence for each divided area based on history information, rather than treating all areas uniformly. This allows the camera to concentrate search efforts in areas with higher probability of subject existence, reducing overall search time while maintaining complete coverage through systematic exploration of all divided areas
Solution Approach 2:
The search strategy is made dynamic by continuously updating the probability of subject existence in each divided area based on accumulated history information. The search order and focus are dynamically adjusted according to the current probability distribution, allowing the system to adapt to changing subject locations and patterns over time while ensuring comprehensive search coverage
Data Source
AI summary
In an image shooting control apparatus, an acquisition unit acquires image data obtained by an image sensor, a detection unit detects a subject from the image data, a map creation unit creates an environmental map, a storage unit stores history information that associates the subject with a divided area in which the subject exists among a plurality of divided areas obtained by dividing the environmental map, a determination unit finds probability of existence of the subject for each divided area based on the history information and determines an order of the plurality of divided areas to be searched for the subject based on the probability of existence, and a control unit controls an orientation of the image sensor so as to shoot the plurality of divided areas in the determined order.