Abstract: Embodiments of the present invention facilitate product defect detection. A computer-implemented method comprises: receiving, by a device operatively coupled to one or more processors, a template image of a normal product; generating, by the device, one or more geometric training parameters for transforming the template image; and transforming, by the device, the template image using the one or more geometric training parameters to generate a transformed image for training a data model, wherein the trained data model being used for aligning the template image and an image under inspection of a product.

Abstract: Embodiments of the present invention facilitate product defect detection. A computer-implemented method comprises: receiving, by a device operatively coupled to one or more processors, a template image of a normal product; generating, by the device, one or more geometric training parameters for transforming the template image; and transforming, by the device, the template image using the one or more geometric training parameters to generate a transformed image for training a data model, wherein the trained data model being used for aligning the template image and an image under inspection of a product.

Abstract: A computer-implemented method is provided for image-based, self-guided object detection. The method includes receiving, by a processor device, a set of images. Each of the images has a respective grid thereon that is labeled regarding a respective object to be detected using grid level label data. The method further includes training, by the processor device, a grid-based object detector using the grid level label data. The method also includes determining, by the processor device, a respective bounding box for the respective object in each of the images, by applying local segmentation to each of the images. The method additionally includes training, by the processor device, a Region-based Convolutional Neural Network (RCNN) for joint object localization and object classification using the respective bounding box for the respective object in each of the images as an input to the RCNN.

Abstract: Scheduling automated visual inspection tasks includes capturing an image of a component to be inspected. A visual inspection model is formed with a model engine as a composite model of utility modules and functional modules to perform visual inspection of the image of the component. An abstract processing workflow of the visual inspection model is derived with a scheduler including dependencies between the utility modules and the functional modules. Performance of each of the functional modules is profiled with the scheduler by testing performance with available hardware resources to produce a performance profile. Parallel instances of each of the functional modules in a branch of the abstract processing workflow are scheduled with the scheduler according to the dependencies and the performance profiles. An indication of defects in the component is produced by processing the visual inspection model according to the scheduled functional modules.

Abstract: Techniques for generating machine learning architectures are provided. A data set is received for training one or more machine learning (ML) models, where the data set comprises labeled exemplars for a plurality of classes. The data set is partitioned into a training set and a testing set. A first ML model is trained using the training set, and a quality of the first ML model with respect to each class of the plurality of classes is evaluated using the testing set. Upon determining that the quality of the first ML model is below a predefined threshold with respect to a first class and a second class of the plurality of classes, a subset of the training set is identified, where each exemplar in the subset corresponds to either the first class or the second class. A second ML model is trained using the subset of the training set.

Abstract: A computer implemented method for surface defect inspection that includes recording an optical image of a surface including a defect; converting the optical image including the defect into a heat map; extracting a region of interest including the defect from the heat map; and comparing the region of interest including the defect from the heat map to a binary classification model using a sliding window based voting mechanism to determine if the defect is greater than or less than a threshold failure value.

Abstract: Embodiments of the present disclosure relate to object defect detection. In an embodiment, a computer-implemented method is disclosed. According to the method, for a test image of at least one part of a target object, a reference image is generated by repeating a periodic pattern detected in the test image, the target object consisting of elements. A differential image is determined by comparing the test image and the reference image. The differential image is superimposed on a predefined grid image to obtain a superimposed image. The grid image comprises grids corresponding to elements of a reference object associated with the target object. The number of defective elements is determined in the at least one part of the target object based on the superimposed image. In other embodiments, a system and a computer program product are disclosed.

Abstract: A method, a device and a computer program product for image processing are proposed. In the method, whether a first image indicates a defect associated with a target object is determined. In response to determining that the first image indicates the defect, a second image absent from the defect is obtained based on the first image. The defect is identified by comparing the first image with the second image. In this way, the defect associated with the target object in the image can be accurately and efficiently identified or segmented.

Abstract: Semiconductor structures are provided. An exemplary semiconductor structure includes a semiconductor substrate having a first region and a second region and a plurality of first fins on the semiconductor substrate in the first region and a plurality of second fins on the semiconductor substrate in the second region. A first oxide layer is on side surfaces of the plurality of first fins; and a second oxide layer is on side surfaces of the second fins. A corner between a top surface and a side surface of each first fin is a first rounded corner. A corner between a top surface and a side surface of each second fin is a second rounded corner. A radius of curvature of the first rounded corner is different from a radius of curvature of the second corner.

Abstract: An apparatus for disassembling an electronic component from a circuit board includes a base, a controller, a positioning assembly, and a heating and suction assembly. The controller is fixed to the base. The positioning assembly is arranged on the base and coupled to the controller. The positioning assembly is configured to be controlled by the controller to position the circuit board at a first predetermined position. The heating and suction assembly is movably arranged on the base and coupled to the controller. The heating and suction assembly is moved to a second predetermined position to heat the electronic component and moved to a third predetermined position to attract and hold the heated electronic component by suction, whereby the electronic component is disassembled from the circuit board.

Abstract: Methods and systems for detecting abnormal connectivity on a product are provided. The methods receive an inspection image of the product including a defect, match the inspection image with a template image of the product to locate a corresponding normal region in the template image and obtain a differential image between the inspection image and the normal region. The method further includes forming a regional mask image from component masks. Each of the component masks includes a binary image of the template image with only one kind of components of the product remaining. The regional mask image is a region in its corresponding component mask that corresponds to the normal region. The method further includes determining, based on a calculation using the differential image and the at least one regional mask image, at least one of: connectivity relationship, connectivity type and connectivity scale of abnormal connectivity on the product.

Abstract: A method of perceptual video coding based on face detection is provided. The method includes calculating a bit allocation scheme for coding a light field video based on a saliency map of the face, calculating an LCU level Lagrange multiplier for coding a light field video based on a saliency map of the face and calculating an LCU level quantization parameter for coding a light field video based on a saliency map of the face.

Abstract: Embodiments facilitating audio source identification are provided. A computer-implemented method comprises: receiving, by a device operatively coupled to one or more processors, an audio signal under inspection; generating, by the device, an image of time-frequency spectrum of low frequency component and high frequency component of the audio signal; and identifying, by the device, a source of the audio signal based on the generated image and one or more patterns of time-frequency spectrum, wherein each of the one or more patterns is corresponding to low frequency feature and high frequency feature of a specific audio source.

Abstract: A computer-implemented method for implementing cognitive state recognition within a telematics system includes determining moving object operation behaviors of an operator of a moving object corresponding to respective select contextual combinations, generating a personalized base characteristic for each select contextual combination, recognizing a cognitive state based on a comparison of actual moving object operation behaviors associated with the select contextual combinations and corresponding ones of the personalized base characteristics, and automatically triggering one or more actions based on the cognitive state.

Abstract: Embodiments facilitating audio source identification are provided. A computer-implemented method comprises: receiving, by a device operatively coupled to one or more processors, an audio signal under inspection; generating, by the device, an image of time-frequency spectrum of low frequency component and high frequency component of the audio signal; and identifying, by the device, a source of the audio signal based on the generated image and one or more patterns of time-frequency spectrum, wherein each of the one or more patterns is corresponding to low frequency feature and high frequency feature of a specific audio source.

Abstract: Embodiments of the present invention facilitate product defect detection. A computer-implemented method comprises: receiving, by a device operatively coupled to one or more processors, a template image of a normal product; generating, by the device, one or more geometric training parameters for transforming the template image; and transforming, by the device, the template image using the one or more geometric training parameters to generate a transformed image for training a data model, wherein the trained data model being used for aligning the template image and an image under inspection of a product.

Abstract: This disclosure provides a method for object detection. The method comprises receiving a user input that specifies one or more first regions and one or more second regions in a template image. The one or more second regions include one or more objects of interest. The method further comprises for each of the one or more first regions discovering a third region in an image under detection corresponding to the first region in the template image by matching the image under detection with the template image. The method further comprises computing a transformation function based on the matching from each of the one or more first regions to its corresponding third region. The method further comprises applying the computed transformation function to the one or more second regions to localize one or more fourth regions in the image under detection for the object detection.

Abstract: Techniques facilitating object classification based on decoupling a background from a foreground of an image are provided. A system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a model that is trained on images that comprise respective backgrounds and respective foregrounds that are interleaved. The model can be trained to detect the respective backgrounds with a defined confidence level. The computer executable components can also comprise an extraction component that employs the model to identify a background of a received image based on the defined confidence level and to decouple a foreground object of the received image based on identification of the background of the received image.

Abstract: A computer-implemented method for implementing cognitive state recognition within a telematics system includes determining moving object operation behaviors of an operator of a moving object corresponding to respective select contextual combinations, generating a personalized base characteristic for each select contextual combination, recognizing a cognitive state based on a comparison of actual moving object operation behaviors associated with the select contextual combinations and corresponding ones of the personalized base characteristics, and automatically triggering one or more actions based on the cognitive state.

Abstract: In one embodiment, a computer-implemented method includes receiving training data including a plurality of records, each record having a plurality of attributes. The training data is horizontally parallelized across two or more processing elements. This horizontal parallelizing includes dividing the training data into two or more subsets of records; assigning each subset of records to a corresponding processing element of the two or more processing elements; transmitting each subset of records to its assigned processing element; and sorting, at the two or more processing elements, the two or more subsets of records to two or more candidate leaves of a decision tree. The output from horizontally parallelizing is converted into input for vertically parallelizing the training data. The training data is vertically parallelized across the two or more processing elements. The decision tree is grown based at least in part on the horizontally parallelizing, the converting, and the vertically parallelizing.