Camera Face Detection for Flexible Main Object Specification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic cameras with face recognition functions struggle to flexibly handle situations where multiple persons are present, often limiting the number of main objects that can be specified and failing to effectively treat unrecognized individuals as main objects, thereby compromising user-friendliness.
Innovation Solution
The integration of an electronic camera with an image pickup device, memory, face detecting section, face recognizing section, and object specifying section, which detects face areas, extracts characterizing points, and specifies main objects based on proximity, distance, or size, allowing for flexible selection and processing of main objects, including those not initially recognized.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the camera limits the number of main objects that can be specified, then the processing complexity is reduced, but the adaptability to handle multiple persons and unrecognized individuals deteriorates
Solution Approach 1:
The camera dynamically determines the number of main objects based on the shooting situation rather than using a fixed limit. The object specifying section automatically adjusts how many detected faces are treated as main objects, allowing the system to adapt between simple scenes (fewer main objects) and complex scenes (more main objects) without requiring manual configuration or overwhelming processing overhead.
2Reliability
If the camera strictly recognizes only registered faces as main objects, then the reliability of face recognition is improved, but the ease of operation and user-friendliness deteriorates due to inability to handle unrecognized individuals
Solution Approach 1:
The system segments the object specification process into two independent stages: face recognition (matching against registered faces) and object specification (determining main objects based on detected faces). This allows the camera to first reliably identify registered faces, then separately determine which detected faces (including unrecognized ones) should be treated as main objects based on additional criteria like position and size, thus maintaining recognition reliability while improving ease of operation.
Solution Approach 2:
The camera performs preliminary face detection and extraction of characterizing points for all detected faces before performing the final main object specification. This preliminary action allows the system to prepare recognition data in advance, then efficiently determine main objects by comparing against registered faces and applying specification criteria, ensuring both reliable recognition and inclusive handling of unrecognized individuals.
3Measurement precision
If the camera uses multiple criteria (position, distance, size) to specify main objects, then the precision of main object specification is improved, but the device complexity increases
Solution Approach 1:
The camera merges multiple existing functions into the object specifying section to achieve precise main object specification without proportionally increasing device complexity. By combining face detection, characterizing point extraction, position analysis, and main object determination into an integrated object specifying section, the system leverages shared hardware and processing resources, reducing the actual complexity increase despite using multiple criteria.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution enhances user-friendliness by enabling the camera to automatically specify and process main objects, including unrecognized individuals, improving the ability to follow user intent and handle complex scenes, while allowing for customized settings and image processing tailored to main objects.
Implementation Method 1
an image pickup device which photo-electrically converts an image of an object into an electric signal and generates an image signal as the electric signal
Data Source
AI summary
An electronic camera includes an image pickup device, a memory, a face detecting section, a face recognizing section, and an object specifying section. The image pickup device photo-electrically converts an image of an object into an electric signal and generates an image signal as the electric signal. The memory has recorded registration data representing characterizing points of faces as recognizing targets. The face detecting section detects face areas in a shooting image plane based on the image signal and extracts characterizing points of faces of objects from the face areas. The face recognizing section determines whether or not the face areas are the recognizing targets based on data of the characterizing points corresponding to the face areas and on the registration data. The object specifying section specifies as a main object an object present on nearest side of the electronic camera of objects as the recognizing targets.