Oblique Aerial Image Bounding Boxes Using Horizon Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for generating bounding boxes from aerial imagery that depict a horizon are ineffective, resulting in inaccurate or unusable geolocalization due to the horizon's presence, which traditional methods fail to account for.

Innovation Solution

A method is introduced to modify images depicting a horizon by adjusting the frame to cut out the horizon, using a visibility radius and tangent line to determine a meaningful distance, thereby generating accurate bounding boxes for geolocalization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional bounding box generation methods are used on aerial imagery depicting a horizon, then the processing is simple and fast, but the geolocalization accuracy becomes inaccurate or unusable

Engineering Contradiction:
Improvegeolocalization accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by detecting the horizon in the aerial image before generating the bounding box. The horizon detection step identifies the horizon line and calculates the visible Earth radius, which is then used to adjust the bounding box generation process. This preliminary detection and calculation ensures that the subsequent bounding box accurately reflects the visible Earth portion, resolving the contradiction between simple processing and accurate geolocalization.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If images depicting horizons are excluded from processing, then geolocalization accuracy is maintained, but resource utilization decreases and unnecessary image capture occurs

Engineering Contradiction:
Improveresource utilizationVSAvoidgeolocalization accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent converts the harmful effect of the horizon (which traditionally makes images unusable) into a beneficial element by using the horizon detection to define the visible Earth radius. The horizon line becomes a reference for calculating the tangent line and determining the appropriate bounding box extent. This approach allows images with horizons to be processed effectively, improving resource utilization while maintaining geolocalization accuracy.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

3Measurement precision

If the visible Earth radius is not considered in bounding box generation, then the processing is simpler, but the bounding box extends beyond the visible Earth portion reducing accuracy

Engineering Contradiction:
Improvebounding box accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary calculation of the visible Earth radius based on camera altitude and the horizon line detection before generating the bounding box. This pre-calculated radius is then used to constrain the bounding box extent, ensuring it does not extend beyond the visible Earth portion. The preliminary action of calculating the visible radius prevents the need for complex iterative adjustments later, balancing accuracy with processing efficiency.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20250245982A1Generating bounding boxes for geolocalizing oblique aerial imagery
Publication Date: 2025.07.31 X DEVELOPMENT LLC
  • US20250245982A1 patent drawing
  • US20250245982A1 patent drawing
  • US20250245982A1 patent drawing

AI summary

Methods, systems, and apparatus for receiving a first image file recording a first image and a first set of metadata associated with the first image, determining that the first image depicts a horizon, and in response, providing a modified first set of metadata by applying a visibility radius to a projection of the Earth depicted in the first image, determining a tangent line based on the visibility radius, and adjusting a height of the first image based on the tangent line to provide a modified height in the modified first set of metadata, and outputting a first geographic features file that is generated using the modified first set of metadata, the first geographic features file including data representing one or more geographic features represented in the first image file.