Abstract: According to embodiments, a method, computer system, and computer program product for obtaining boundaries of structural defects of materials in images of structures is provided. The present invention may include loading an image of a structure made of a material having one or more structural defects and running a pipeline according to certain processing parameters. Running the pipeline pre-processes the loaded image to obtain an initial segmentation mask, where the mask defines an initial boundary of the crack. Based on the initial segmentation mask obtained, a graph of a skeletal structure of the crack is generated, where the skeletal structure comprises a backbone and outer substructures. The graph is pruned by cutting away one or more outer subgraphs corresponding to respective outer substructures to obtain a revised skeletal structure. A revised boundary of the crack is obtained based on both the loaded image and the revised skeletal structure.

Abstract: A computer-implemented method is disclosed for image stitching for a plurality of digital images of an infrastructure surface for defect detection. The method includes providing a plurality of partial digital images of an infrastructure surface. The method includes extracting global positioning system meta-data from data corresponding to the partial digital images. The method includes determining feature descriptions of features in the one or more partial digital images. The method includes executing a scheduled processing sequence for the partial digital images based on the extracted global positioning system meta-data including determining an affinity matrix using the feature descriptions of adjacent partial digital images to incrementally position each of the partial digital images such that an overview image of the infrastructure surface is produced by iteratively digitally stitching the plurality of partial digital images together.

Abstract: According to embodiments, a method, computer system, and computer program product for obtaining boundaries of structural defects of materials in images of structures is provided. The present invention may include loading an image of a structure made of a material having one or more structural defects and running a pipeline according to certain processing parameters. Running the pipeline pre-processes the loaded image to obtain an initial segmentation mask, where the mask defines an initial boundary of the crack. Based on the initial segmentation mask obtained, a graph of a skeletal structure of the crack is generated, where the skeletal structure comprises a backbone and outer substructures. The graph is pruned by cutting away one or more outer subgraphs corresponding to respective outer substructures to obtain a revised skeletal structure. A revised boundary of the crack is obtained based on both the loaded image and the revised skeletal structure.

Abstract: Aspects of the present invention disclose a method, computer program product, and system for improving object detection in high-resolution images at inference time. The method includes one or more processors receiving a high-resolution image. The method further includes one or more processors decomposing the received image into hierarchically organized layers of images. Each layer comprises at least one image tile of the received image. Each of the image tiles have a corresponding resolution suitable to a baseline image recognition algorithm. The method further includes one or more processors applying the baseline algorithm to each of the image tiles of each layer. The method further includes one or more processors performing a result aggregation of results of the baseline algorithm applications to the image tiles of the layers.

Abstract: A computer-implemented method is disclosed for image stitching for a plurality of digital images of an infrastructure surface for defect detection. The method includes providing a plurality of partial digital images of an infrastructure surface. The method includes extracting global positioning system meta-data from data corresponding to the partial digital images. The method includes determining feature descriptions of features in the one or more partial digital images. The method includes executing a scheduled processing sequence for the partial digital images based on the extracted global positioning system meta-data including determining an affinity matrix using the feature descriptions of adjacent partial digital images to incrementally position each of the partial digital images such that an overview image of the infrastructure surface is produced by iteratively digitally stitching the plurality of partial digital images together.

Abstract: Aspects of the present invention disclose a method, computer program product, and system for improving object detection in high-resolution images at inference time. The method includes one or more processors receiving a high-resolution image. The method further includes one or more processors decomposing the received image into hierarchically organized layers of images. Each layer comprises at least one image tile of the received image. Each of the image tiles have a corresponding resolution suitable to a baseline image recognition algorithm. The method further includes one or more processors applying the baseline algorithm to each of the image tiles of each layer. The method further includes one or more processors performing a result aggregation of results of the baseline algorithm applications to the image tiles of the layers.

Abstract: Determining cognitive models to be deployed at auxiliary devices may include maintaining relations, e.g., in a database. The relations map hardware characteristics of auxiliary devices and example datasets to cognitive models. Cognitive models are determined for auxiliary devices, based on said relations, e.g., for each of the auxiliary devices. An input dataset is accessed, which comprises data of interest, e.g., collected at a core computing system (CCS), and hardware characteristics of each of the auxiliary devices. An auxiliary cognitive model is determined based on a core cognitive model run on the input dataset accessed, wherein the core cognitive model has been trained to learn at least part of said relations. Parameters of the auxiliary model determined can be communicated to said each of the auxiliary devices for the latter to deploy the auxiliary model determined. Method may be implemented in a network having an edge computing architecture.

Abstract: The present disclosure relates to a hardware accelerator for executing a computation task composed of a set of operations. The hardware accelerator comprises a controller and a set of computation units. Each computation unit of the set of computation units is configured to receive input data of an operation of the set of operations and to perform the operation, wherein the input data is represented with a distinct bit length associated with each computation unit. The controller is configured to receive the input data represented with a certain bit length of the bit lengths and to select one of the set of computation units that can deliver a valid result and that is associated with a bit length smaller than or equal to the certain bit length.

Abstract: Determining cognitive models to be deployed at auxiliary devices may include maintaining relations, e.g., in a database. The relations map hardware characteristics of auxiliary devices and example datasets to cognitive models. Cognitive models are determined for auxiliary devices, based on said relations, e.g., for each of the auxiliary devices. An input dataset is accessed, which comprises data of interest, e.g., collected at a core computing system (CCS), and hardware characteristics of each of the auxiliary devices. An auxiliary cognitive model is determined based on a core cognitive model run on the input dataset accessed, wherein the core cognitive model has been trained to learn at least part of said relations. Parameters of the auxiliary model determined can be communicated to said each of the auxiliary devices for the latter to deploy the auxiliary model determined. Method may be implemented in a network having an edge computing architecture.

Abstract: A machine-learning model generation method, system, and computer program product deciding, via a first algorithm, a machine-learning algorithm that is best for customer data, invoking the machine-learning algorithm to train a neural network model with the customer data, analyzing the neural network model produced by the training for an accuracy, and improving the accuracy by iteratively repeating the training of the neural network model until a customer-defined constraint is met, as determined by the first algorithm.

Abstract: The disclosure provides an approach for edge-aware spatio-temporal filtering. In one embodiment, a filtering application receives as input a guiding video sequence and video sequence(s) from additional channel(s). The filtering application estimates a sparse optical flow from the guiding video sequence using a novel binary feature descriptor integrated into the Coarse-to-fine PatchMatch method to compute a quasi-dense nearest neighbor field. The filtering application then performs spatial edge-aware filtering of the sparse optical flow (to obtain a dense flow) and the additional channel(s), using an efficient evaluation of the permeability filter with only two scan-line passes per iteration. Further, the filtering application performs temporal filtering of the optical flow using an infinite impulse response filter that only requires one filter state updated based on new guiding video sequence video frames.

Abstract: The disclosure provides an approach for edge-aware spatio-temporal filtering. In one embodiment, a filtering application receives as input a guiding video sequence and video sequence(s) from additional channel(s). The filtering application estimates a sparse optical flow from the guiding video sequence using a novel binary feature descriptor integrated into the Coarse-to-fine PatchMatch method to compute a quasi-dense nearest neighbor field. The filtering application then performs spatial edge-aware filtering of the sparse optical flow (to obtain a dense flow) and the additional channel(s), using an efficient evaluation of the permeability filter with only two scan-line passes per iteration. Further, the filtering application performs temporal filtering of the optical flow using an infinite impulse response filter that only requires one filter state updated based on new guiding video sequence video frames.