An artificial neural network-based method for detecting a surface type of an object includes: receiving a plurality of object images, wherein a plurality of spectra of the plurality of object images are different from one another and each of the object images has one of the spectra; transforming each object image into a matrix, wherein the matrix has a channel value that represents the spectrum of the corresponding object image; and executing a deep learning program by using the matrices to build a predictive model for identifying a target surface type of the object. Accordingly, the speed of identifying the target surface type of the object is increased, further improving the product yield of the object.

A system for acquiring three-dimensional (3-D) models of objects includes a first camera group including: a first plurality of depth cameras having overlapping fields of view; a first processor; and a first memory storing instructions that, when executed by the first processor, cause the first processor to: control the first depth cameras to simultaneously capture a first group of images of a first portion of a first object; compute a partial 3-D model representing the first portion of the first object; and detect defects in the first object based on the partial 3-D model representing the first portion of the first object.

Various examples include a system and network to map of substrates within a substrate carrier (e.g., such as silicon wafers within a wafer cassette), and a classification of a state of each substrate, as well as the carrier in which the substrates are placed. In various examples provided herein, an image acquisition system, such as a camera, acquires multiple images of the substrates within the carrier. The image or images are then processed with a deep-convolutional neural-network to classify a state of the substrate relative to a substrate slot including empty slots, occupied slots (e.g., properly loaded slots), double-loaded slots, cross-slotted, and protruded (where a substrate is not fully loaded into a slot).

The present disclosure provides techniques for optical inspection systems and methods for moving objects. In some embodiments, an optical inspection system includes: a first and a second image capturing device configured to acquire images from moving objects; a first and a second first-stage storage system; a first and a second second-stage processor; a second-stage storage system; a third-stage processor; and a third-stage storage system. In some embodiments, an optical inspection system, includes: a first and a second image capturing device; a first and a second volatile memory system; a first and a second second-stage processor; a third second-stage processor; and a third-stage storage system. The first and second second-stage processors can be configured to analyze the images from the image capturing devices. The third-stage processor or the third second-stage processor can be configured to process information from a processor and/or storage system and produce a report.

A system and a method for assessing reliability of an electronic component. The method may include training a machine earning (ML) algorithm and/or a classification network to classify electronic components based on one or more features, attributes or characteristics related to reliability of the electronic components, e.g., related to a level of solderability of the components lead or balls or features indicating of tampering of the electronic component. By receiving an image of a test electronic component and extracting a feature related to reliability of the test electronic component from the image received, embodiments of the invention may enable classifying the test electronic component to a class indicating a reliability of the test electronic component by using the machine learning algorithm and/or the classification network.

A computer-implemented method for generating a training dataset for training an artificial neural network configured to use images of lateral faces of a timber board to provide information about structure and/or defects, the method including; a log generation step during which a virtual model of a log is generated; a sawing step of the virtual model to obtain one or more virtual timber boards; a pattern step during which a surface pattern is determined as the intersection between the virtual lateral face and the internal structure and/or defects; a rendering step during which a rendered surface image of the lateral face of the virtual timber board is created; and an input data generation step during which the rendered surface images are used to create one or more item of input data; an output data generation step during which an item of output data is generated; and a population step during which a record is added to the training dataset comprising the item of input data, in combination with the item of output data.

A training device and a training method for a neural network model. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.

Devices, systems and methods for sorting and labeling food products are provided. Respective spectra of food products for a plurality of segments of a line are received at a controller from at least one line-scan dispersive spectrometer configured to acquire respective spectra of the food products for the plurality of segments of the line. The controller applies one or more machine learning algorithms to the respective spectra to classify the plurality of segments according to at least one of one or more food parameters. The controller controls one or more of a sorting device and a labeling device according to classifying the plurality of segments to cause the food products to be one or more of sorted and labeled according to the at least one of the one or more food parameters.

A method detects mechanical equipment parts. The method includes: obtaining an image of a part; extracting a feature from the image using a machine learning model, identifying a type of surface defect on the basis of the feature to obtain an identification result; and determining whether to replace the part on the basis of the identification result and a predetermined standard of the part. The method reduces the difficulty of detecting a part, can accurately identify a surface defect of the part and determine whether the part needs to be replaced, thereby improving the work efficiency, and shortens the time for mechanical equipment to stop operating for maintenance, thus improving the operating efficiency of the mechanical equipment. The method is automatically executed by a computer, thereby avoiding manually checking errors, improving the accuracy of detection results, and thus improving the reliability of operation of the mechanical equipment.

A method for inspection for a photovoltaic module or cell is disclosed. The method includes acquiring one or more polarimetric images of the photovoltaic module or cell using a camera which may include a polarization sensor, analyzing the one or more polarimetric images, and identifying a presence of a defect in the photovoltaic module or cell. A device for inspection for a photovoltaic module or cell is also disclosed, wherein the device includes a camera having a polarimetric sensor and is configured to be positioned at one or more locations relative to a location of the photovoltaic module or cell.