Image Processing Apparatus for Antenna Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing technologies face challenges in processing load and accuracy when detecting predetermined objects in images, particularly in distinguishing antenna devices from outdoor backgrounds.
Innovation Solution
An information processing apparatus that includes an image acquisition unit for acquiring annotated images, an adjustment image generation unit for parameter adjustment, and a machine learning unit for generating a learning model to detect objects, along with an angle calculation unit to determine object orientation, improves detection accuracy and efficiency by correcting annotations and expanding learning data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing technologies are used to detect predetermined objects, then the processing can be performed, but the processing load is high and the detection accuracy is insufficient
Solution Approach 1:
The patent applies preliminary action by pre-adjusting image parameters (brightness, contrast, saturation) before feeding images to the machine learning model. This preprocessing step enhances the visual characteristics of antenna devices in advance, enabling the model to detect and distinguish them more accurately from outdoor backgrounds without increasing computational load during the detection phase.
Solution Approach 2:
The patent changes image parameters such as brightness, contrast, and saturation to optimize the visual representation of antenna devices. By adjusting these parameters, the system enhances the distinguishability of antenna devices from their background, thereby improving detection accuracy while maintaining efficient processing speeds.
2Measurement precision
If conventional technologies are used to distinguish antenna devices from outdoor backgrounds, then processing can be performed, but the detection accuracy is insufficient
Solution Approach 1:
The patent applies color changes by adjusting the saturation and brightness parameters of the image. This enhances the visual contrast between antenna devices and outdoor backgrounds, making it easier for the machine learning model to distinguish and detect antenna devices accurately based on their color and luminance characteristics.
Solution Approach 2:
The system performs preliminary parameter adjustment on images before they are input to the detection model. This pre-processing enhances the visual characteristics of antenna devices in advance, enabling more accurate distinction from backgrounds during the detection phase without adding computational complexity.
3Reliability
If machine learning is performed using annotated images, then a learning model can be generated, but the annotation process requires high processing load and time
Solution Approach 1:
The patent adjusts image parameters to enhance the visibility and distinguishability of antenna devices before annotation. This makes the annotation process more efficient by clearly highlighting the objects that need labeling, thereby reducing the time and effort required for accurate annotation while maintaining high learning model quality.
Solution Approach 2:
By modifying color and brightness parameters, the system enhances the visual characteristics of antenna devices in the annotated images. This improves the quality of annotations by making objects more distinct, thereby reducing the time required for accurate labeling while maintaining reliable learning model output.
Data Source
AI summary
[Problem] Provided is a novel information processing technology relating to a predetermined object in an image.[Solving Means] An information processing apparatus includes: an image acquisition unit that acquires an image used as teacher data for machine learning, the image being with one or a plurality of annotations for showing a position in the image at which a predetermined object is shown; an adjustment image generation unit that generates an adjustment image in which parameters of the image are adjusted; and a machine learning unit that generates a learning model for detecting the predetermined object in an image by performing machine learning using teacher data including the adjustment image.