Image Processing Device Face Region Scene Judgment Tone Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods may erroneously detect main images and judge scenes when the reference color is present outside the intended object, leading to incorrect detections and judgments.
Innovation Solution
An image processing device that detects face regions and judges scenes based on pixel values of face and non-face regions, performing tone correction according to the judged scene to enhance image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If color density determination is performed using a reference color selected from an object with high likelihood of being the main image, then main image detection can be performed, but erroneous main image detection occurs when the reference color is present in portions other than the intended object
Solution Approach 1:
The patent divides the image into face region and nonface region, and further segments the nonface region into background region and other region. This segmentation allows the system to distinguish between different types of regions and apply appropriate processing to each, preventing erroneous detection when reference color appears in non-face areas.
Solution Approach 2:
The patent introduces a skin color determination unit as an intermediary that first determines whether a detected region is a face region by checking if it contains skin-colored portions. This intermediary step prevents direct color density determination from producing erroneous results when reference color appears in non-face regions.
2Productivity
If scene judgment is performed based on color density determination, then scene identification can be achieved, but incorrect scene judgment occurs when reference color is present outside the intended object
Solution Approach 1:
The patent segments the image into face region and nonface region, and further divides the nonface region into background region and other region. This segmentation enables accurate scene judgment by considering the spatial relationship between face region and other regions, preventing incorrect judgments when reference color appears in non-face areas.
Solution Approach 2:
The patent introduces a face region determination unit as an intermediary that first identifies face regions before performing scene judgment. This intermediary step ensures that scene judgment is based on accurate identification of face regions, preventing incorrect scene judgment when reference color appears in non-face regions.
3Ease of manufacture
If tone correction is performed without scene-based parameter adjustment, then image processing can be completed, but image quality is not optimized for different scene types
Solution Approach 1:
The patent makes the tone correction parameters dynamic by adjusting them according to the determined scene type (normal scene, backlit scene, night scene). The parameter setting unit selects different parameter values based on scene determination, allowing the tone correction unit to optimize image quality for each specific scene type rather than using fixed parameters.
Solution Approach 2:
The patent changes the tone correction parameters based on scene type determination. The parameter setting unit sets different parameter values for normal scenes, backlit scenes, and night scenes, allowing the tone correction unit to adjust image brightness and contrast appropriately for each scene type, thereby optimizing image quality.
Data Source
AI summary
An image processing device includes: a face detection unit that detects a face region of a person from an image formed by a plurality of pixels; and a scene judgment unit that judges a scene of the image according to a pixel value of the face region of the person detected by the face detection unit and a pixel value of a nonface region excluding the face region in the image.