Abstract: A method and system are provided for assessing damage to a structure. According to one embodiment, the method includes using a classifier that is trained to associate respective portions of in an image of a structure with respective external parts of the structure to detect several of external parts of a structure in a first image of the structure. The method also includes using a first machine learning system, trained to perform segmentation of an image, to identify one or more image segments in a second image, where each image segment represents damage of a particular type and, using a second machine learning system, trained to associate image segments with external parts identified in an image, to associate a first image segment with a first external part, indicating damage to the first external part.

Abstract: Methods and computing apparatus are provided for training AI/ML systems and use of such systems for performing image analysis so that the damaged parts of a physical structure can be identified accurately and efficiently. According to one embodiment, a method includes selecting an AI/ML system of a particular type; and training the AI/ML system using a dataset comprising one or more auto-labeled images. The auto-labeling was performed using the selected AI/ML system configured using a parts-identification model. The configuration of the trained AI/ML system is output as an improved parts-identification model.

Abstract: A method and system are provided for assessing damage to a structure. According to one embodiment, the method includes using a classifier that is trained to associate respective portions of in an image of a structure with respective external parts of the structure to detect several of external parts of a structure in a first image of the structure. The method also includes using a first machine learning system, trained to perform segmentation of an image, to identify one or more image segments in a second image, where each image segment represents damage of a particular type and, using a second machine learning system, trained to associate image segments with external parts identified in an image, to associate a first image segment with a first external part, indicating damage to the first external part.