Platform, Systems, And Methods For Identifying Property Characteristics And Property Feature Conditions Through Imagery Analysis
Patent No. US11030491 (titled "Platform, Systems, And Methods For Identifying Property Characteristics And Property Feature Conditions Through Imagery Analysis") was filed by Aon Re Inc on Jan 22, 2021.
What is this patent about?
’491 is related to the field of automated property assessment using image analysis. Specifically, it addresses the problem of efficiently and accurately determining the condition of various features of a property, such as the roof, siding, or surrounding structures, using machine learning techniques applied to aerial and terrestrial imagery. This information is valuable for insurance risk assessment, property valuation, and maintenance planning.
The underlying idea behind ’491 is to leverage machine learning to automate the process of identifying property features and assessing their condition from images. The system uses a combination of image processing techniques and machine learning models to first identify the features of interest (e.g., roof shape, presence of a pool) and then classify their condition (e.g., good, fair, poor) based on visual cues in the images.
The claims of ’491 focus on a system and method for automatically assessing features of a property location. This involves accessing multiple images, applying boundary information to isolate the property, and classifying the condition of property features. The classification process uses machine learning algorithms to determine characteristics of the feature and then classify its condition based on those characteristics.
In practice, the system would take aerial or street-view images of a property and use image processing techniques to identify the boundaries of the property and the features of interest. Then, it would apply machine learning models, such as convolutional neural networks , trained on large datasets of labeled images, to classify the condition of each feature. For example, the system might identify the roof shape and then classify its condition as good, fair, or poor based on the presence of damage, wear, or other visual cues.
This approach differs from traditional methods that rely on manual inspections or simple rule-based image analysis. By using machine learning, the system can automatically learn to identify and classify a wide range of property features and conditions with high accuracy. The use of multiple images and boundary information further improves accuracy by allowing the system to focus on the relevant areas of the images and to account for variations in image quality and perspective.
How does this patent fit in bigger picture?
Technical landscape at the time
In the mid-2010s when ’491 was filed, deep learning was increasingly being applied to image recognition tasks, at a time when systems commonly relied on complex feature engineering rather than end-to-end training. Extracting meaningful information from aerial imagery was non-trivial, and often required specialized hardware or software to process the large datasets involved.
Novelty and Inventive Step
The examiner allowed the application because a terminal disclaimer was filed and accepted. This removed a double patenting rejection. The examiner stated that the reasons for allowance were similar to those in a previous notice of allowance for application 15/714,376, and that the previous office action was void of prior art rejection.
Claims
This patent contains 20 claims, with claims 1 and 11 being independent. Independent claim 1 focuses on a system for automatically assessing features of a property location using image analysis and machine learning. Independent claim 11 focuses on a method for automatically assessing risk damage to a property, also using image analysis and machine learning. The dependent claims generally elaborate on and refine the specifics of the system and method described in the independent claims, adding details regarding boundary information, image selection, property characteristics, risk profiles, and user interfaces.
Key Claim Terms New
Definitions of key terms used in the patent claims.
Litigation Cases New
US Latest litigation cases involving this patent.
Patent Family
File Wrapper
The dossier documents provide a comprehensive record of the patent's prosecution history - including filings, correspondence, and decisions made by patent offices - and are crucial for understanding the patent's legal journey and any challenges it may have faced during examination.
Date
Description
Get instant alerts for new documents
US11030491
- Application Number
- US17155633
- Filing Date
- Jan 22, 2021
- Status
- Granted
- Expiry Date
- Sep 25, 2037
- External Links
- Slate, USPTO, Google Patents