UI Element Detection Using Combined Selector, CV, and OCR Attributes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current graphical element detection techniques for robotic process automation (RPA) in user interfaces (UIs) are not optimal when used individually, as they fail to accurately identify UI elements in diverse scenarios, particularly with varying visual appearances.

Innovation Solution

A computer-implemented method that combines UI descriptor attributes from multiple graphical element detection techniques, such as selectors, computer vision (CV), and optical character recognition (OCR), to enhance the accuracy of UI element detection by using a combination of attributes for exact or threshold matches.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single graphical element detection technique is used, then the device complexity is reduced, but the measurement precision of UI element detection deteriorates

Engineering Contradiction:
ImproveUI element detection accuracyVSAvoiddetection technique complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple graphical element detection techniques (selectors, computer vision, OCR) into a unified detection system. The system merges these techniques by collecting UI descriptor attributes from all techniques and selecting the most accurate match based on predefined accuracy thresholds, thereby achieving higher detection accuracy without requiring complex individual techniques to function independently

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The invention creates a universal detection framework that can handle multiple UI element types and scenarios using a single integrated approach. The system universally applies multiple detection techniques to various UI elements and automatically selects the appropriate technique or combination based on the element's characteristics and the detection accuracy requirements

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

2Measurement precision

If multiple graphical element detection techniques are used individually, then the measurement precision improves for diverse scenarios, but the device complexity increases

Engineering Contradiction:
ImproveUI element detection accuracyVSAvoiddetection technique complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges multiple detection techniques into a single coordinated system that collects attributes from selectors, computer vision, and OCR techniques. The system then processes these attributes together and selects the most accurate match based on accuracy thresholds, achieving high precision for diverse scenarios while managing complexity through unified processing

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The invention segments the detection process into distinct phases: collecting UI descriptor attributes from multiple techniques, processing and comparing these attributes, and selecting the final match based on accuracy thresholds. This segmentation allows each technique to contribute its strengths while the overall system maintains manageable complexity through structured processing

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20220050584A1Graphical element detection using a combination of user interface descriptor attributes from two or more graphical element detection techniques
Publication Date: 2022.02.17 UIPATH INC
  • US20220050584A1 patent drawing
  • US20220050584A1 patent drawing
  • US20220050584A1 patent drawing

AI summary

Graphical element detection using a combination of user interface (UI) descriptor attributes from two or more graphical element detection techniques is disclosed. UI descriptors may be used to compare attributes for a given UI descriptor with attributes of UI elements found at runtime in the UI. At runtime, the attributes for the UI elements found in the UI can be searched for matches with attributes for a respective RPA workflow activity, and if an exact match or a match within a matching threshold is found, the UI element may be identified and interacted with accordingly.