ATR Spot Classification for Retroreflective Survey Targets

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing geodetic surveying devices face challenges in robustly distinguishing desired retroreflective targets like prisms or reflective tapes from undesired objects such as windows or vests, particularly under unfavorable environmental conditions, leading to misclassification and delays in the surveying process.

Innovation Solution

A geodetic surveying device equipped with an ATR-illuminator and ATR-sensor, along with a classification model, automatically classifies retroreflective targets by analyzing spot characteristics using machine learning techniques, allowing for real-time differentiation between desired and undesired targets without interrupting the surveying process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional return signal strength threshold classification is used, then the device can operate with simple classification logic, but it misclassifies strongly reflecting foreign objects (windows, warning vests) as targets and fails to detect targets under unfavorable conditions

Engineering Contradiction:
Improvetarget classification accuracyVSAvoidclassification system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent transitions from one-dimensional classification (return signal strength only) to two-dimensional classification by introducing spot shape analysis as a second dimension. The classification model evaluates both the intensity and the spatial distribution of reflected light spots to distinguish between valid targets and foreign objects, thereby improving reliability without excessive complexity increase

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent changes the classification parameters from simple signal strength thresholds to a combination of spot characteristics including intensity, shape, and size. This parameter transformation enables the system to differentiate between targets with similar reflectivity but different geometric properties, resolving the misclassification issue

Inventive Principle:
Principle #35Parameter changes

2Reliability

If simple return signal threshold classification is used, then the device structure remains simple, but targets become invisible under unfavorable conditions (rain, fog, darkness, dirt) as signals fall below threshold or are lost in noise

Engineering Contradiction:
Improvetarget detection robustnessVSAvoidtarget detection difficulty
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent adds spot shape analysis as an additional dimension to target detection, enabling the system to identify targets even when signal strength is reduced by environmental factors. By analyzing the geometric characteristics of the spot in addition to its intensity, the system can distinguish valid targets from noise and weak signals caused by adverse conditions

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If ATR-illuminator and classification model are added to improve target classification, then classification accuracy improves, but device complexity increases

Engineering Contradiction:
Improvetarget classification accuracyVSAvoidsystem component complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The ATR-illuminator serves multiple functions: it provides illumination for the ATR-sensor, enables spot shape analysis, and works in conjunction with the classification model to achieve accurate target identification. This multi-functionality justifies the added complexity by delivering significant performance improvements across multiple operational dimensions

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

Enables accurate and efficient classification of retroreflective targets, ensuring safer and more time-efficient surveying by automatically distinguishing desired targets from undesired ones, even in challenging conditions, thereby enhancing user safety and surveying efficiency.

Implementation Method 1

configured to emit a radiation beam, in particular in a beam divergence of 0.5° to 5°, an target recognition sensor, called ATR-sensor, defining a field-of-view corresponding to the beam divergence, and having a position-sensitive area for receiving radiation of the radiation beam reflected at retroreflective targets

Methodology Applied
Scientific EffectRetroreflection: Retroreflector

Implementation Method 2

a telescope, wherein the telescope is configured for carrying out a distance measurement by means of a laser beam emitted via a beam exit of the telescope

Methodology Applied
Scientific EffectLaser: Laser

Data Source

PatentUS12467750B2Surveying device for improved target classification
Publication Date: 2025.11.11 LEICA GEOSYSTEMS AG
  • US12467750B2 patent drawing
  • US12467750B2 patent drawing
  • US12467750B2 patent drawing

AI summary

A geodetic surveying device, wherein the geodetic surveying device is configured for surveying retroreflective cooperative targets, the geodetic surveying device comprising a base, a telescope and a support. The geodetic surveying device further comprises a target recognition emitting unit, called ATR-illuminator, an target recognition sensor, called ATR-sensor, an angle encoder and a processing unit. The geodetic surveying device is configured to distinguish desired targets from undesired targets, wherein the geodetic surveying device further comprises a classification model, wherein the classification model is configured for classification of retroreflective targets generating spots, wherein the retroreflective targets are classified into desired targets and undesired targets, wherein the classification is carried out by applying the classification model on the spot.