Subject Detection Model Update via Image Stabilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image detection methods face challenges in accurately detecting subjects when images are shaky, leading to blurred images and inaccurate subject identification.
Innovation Solution
The method involves obtaining a current and previous image, determining if the current image indicates a shaking state, calculating a transformation matrix, correcting the previous image based on this matrix, and updating a subject detection model to improve the accuracy of subject detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subject detection is performed directly on shaky images, then the detection process is simple and fast, but the subject identification accuracy deteriorates due to blurred images
Solution Approach 1:
The patent applies preliminary action by performing image stabilization through transformation matrix calculation and image correction before subject detection is executed. The system calculates the transformation matrix from shaking detection results, corrects the shaky image to generate a stabilized image, and then performs subject detection on the corrected image, thereby improving detection accuracy before the main detection task occurs.
Solution Approach 2:
The patent introduces an intermediary element by using a correction image as a mediator between the original shaky image and the final subject detection process. The correction image, generated through transformation matrix application, serves as an intermediate stabilized version that bridges the gap between the shaky input and the detection algorithm, improving subject identification without requiring direct processing of the unstable original image.
2Stability of the object's composition
If image correction based on transformation matrix is performed, then image stability improves, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary shaking detection and transformation matrix calculation before the main image correction and subject detection tasks. By pre-calculating the transformation matrix from shaking detection results and applying it to correct the image beforehand, the system stabilizes the image in advance, allowing subsequent detection to proceed more efficiently on the already-stabilized correction image.
Data Source
AI summary
The present disclosure relates to a method and an apparatus for detecting a subject, an electronic device, and a computer readable storage medium. The method includes the following. A current image and a previous image are obtained. A transformation matrix between the current image and the previous image is obtained in response to determining that the current image indicates the shaking state. The previous image is corrected based on the transformation matrix. The subject detection model is updated based on the corrected previous image. The subject detection is performed on the current image based on the updated subject detection model, to obtain a target subject.