Abstract: In an example embodiment, rather than a single calibration object, a plurality of calibration objects are placed within the field of view of an inspection camera at the same time. Each of the calibration objects is or includes a semi-spherical reflective object having known sizes. During a calibration process, the distance between a light source and camera may be determined by activating the light source and capturing an image of how the light source bounces off each of the calibration objects, and more particularly the location on each of the calibration objects that the light source appears to reflect from. This allows the angle that the light source strikes each calibration object to be calculated, and these angles can then be used to determine the distance between the light source and the camera.

Abstract: In an example, a light system is provided that allows for increased reliability in directly detecting defects in components by capturing multiple different images of the component under different lighting conditions and using a photometric stereo technique to reconstruct a three-dimensional (3D) model of the component, which can be used to perform additional operations to increase the reliability of defect detection, such as measuring the height of variations in a surface of the component.

Abstract: A feature inspection lighting system comprises a body including an inner surface, the inner surface having facets defined therein. A PCB substrate formed of a ductile material is mounted to the inner surface of the body. The PCB substrate comprises segments corresponding to the facets of the inner surface of the body. Light emitters are mounted to the PCB substrate. The thermal interface between the segments of the PCB substrate and the facets of the body facilitates heat transfer. A flexible PC electrically connects the PCB substrate to a cable for supplying power to the lighting system. One or more diffusers are provided to achieve the desired illumination of a target object and to seal the interior of the assembly from contaminants.

Abstract: In an example embodiment, two different neural networks are used in conjunction with each other to identify defects in items appearing in images. More particularly, a defect neural network is trained on images labeled with an indication of a defect type (or no defect) appears in an item in the image (with optionally an identification of the area in which the defect appears). Separately, a location neural network is trained on images that have a specific geometric shape (e.g., point, line, circle) labeled in the image. The labels identify not just the pixels that are part of the shape, but also provide a value that indicates a degree of the identification of the location. Surrounding pixels to the identified shapes are then given values according to a distribution that is based on distance from the shape, with the peak values being those pixels that are exactly on the identified shape.

Abstract: In an example embodiment, the controller is designed with a single current sensing circuit that is able to measure the current on multiple LED channels, eliminating the need for each LED channel to have its own current sensing circuit. The current sensing circuit may further be utilized with a control unit, which has an LED control model selection multiplexor (MUX) that switches between real-time closed-loop control and open-loop control.

Abstract: In an example embodiment, a controller and computer system are provided for a lighting apparatus. More particularly, users are able to use the computer system to define a sequence of one or more capture configurations. Each capture configuration includes identification(s) of one or more light sources to illuminate during an image capture, one or more channels to use to capture the image, and an exposure time. Each sequence can further define custom delays between capture configurations within the sequence. Upon receipt of a trigger from either hardware or software, the controller then fires the stored sequence, which takes images based on the captured configurations (and delays). Artificial intelligence techniques can then be applied to the captured images to identify suspected defects in components captured in the images.

Abstract: In an example embodiment, the controller is designed with a single current sensing circuit that is able to measure the current on multiple LED channels, eliminating the need for each LED channel to have its own current sensing circuit. The current sensing circuit may further be utilized with a control unit, which has an LED control model selection multiplexor (MUX) that switches between real-time closed-loop control and open-loop control.

Abstract: A feature inspection lighting system includes an outer cover having a mounting structure at a first end thereof and defining an outer periphery at a second end thereof. The cover includes an inner surface between the first end and the second end. The inner surface has raised surfaces defined thereon. A plurality of light sources is coupled to the raised surfaces of the inner surface of the cover, and cables to provide electrical power to the light sources is routed between the raised surfaces of the outer cover. The outer periphery of the cover may have a cutout defined therein through which all or part of an object to be inspected can pass.

Abstract: The present disclosure provides a dimensional measurement method and device based on deep learning. The method includes capturing a target image of a target object, obtaining measurement data for the target image, and determining whether or not the target object is within a preset tolerance.

Abstract: The present disclosure provides a dimensional measurement method and device based on deep learning. The method includes capturing a target image of a target object, obtaining measurement data for the target image, and determining whether or not the target object is within a preset tolerance.

Abstract: A fused imaging device and method are disclosed. The device comprises a light source component, an image capture component, and a control component. The control component may be configured to control the light source component to illuminate a target object based on a preset plurality of first optimal lighting configurations. The control component may further control the image capture component to capture images of the target object to obtain multiple first images, under the illumination of the light source component and generate a target image of the target object, based on the first images. The control component may further adjust the incidence angle, pattern, and wavelength of the light source in the light source component, as well as the exposure, lens focus, and polarization of the image capture component, to detect target object defects in captured images of the target object.

Abstract: A fused imaging device and method are disclosed. The device comprises a light source component, an image capture component, and a control component. The control component may be configured to control the light source component to illuminate a target object based on a preset plurality of first optimal lighting configurations. The control component may further control the image capture component to capture images of the target object to obtain multiple first images, under the illumination of the light source component and generate a target image of the target object, based on the first images. The control component may further adjust the incidence angle, pattern, and wavelength of the light source in the light source component, as well as the exposure, lens focus, and polarization of the image capture component, to detect target object defects in captured images of the target object.

Abstract: The present disclosure provides a dimensional measurement method and device based on deep learning. The method includes capturing a target image of a target object, obtaining measurement data for the target image, and determining whether or not the target object is within a preset tolerance.

Abstract: A fused imaging device and method are disclosed. The device comprises a light source component, an image capture component, and a control component. The control component may be configured to control the light source component to illuminate a target object based on a preset plurality of first optimal lighting configurations. The control component may further control the image capture component to capture images of the target object to obtain multiple first images, under the illumination of the light source component and generate a target image of the target object, based on the first images. The control component may further adjust the incidence angle, pattern, and wavelength of the light source in the light source component, as well as the exposure, lens focus, and polarization of the image capture component, to detect target object defects in captured images of the target object.

Abstract: A fused imaging device and method are disclosed. The device comprises a light source component, an image capture component, and a control component. The control component may be configured to control the light source component to illuminate a target object based on a preset plurality of first optimal lighting configurations. The control component may further control the image capture component to capture images of the target object to obtain multiple first images, under the illumination of the light source component and generate a target image of the target object, based on the first images. The control component may further adjust the incidence angle, pattern, and wavelength of the light source in the light source component, as well as the exposure, lens focus, and polarization of the image capture component, to detect target object defects in captured images of the target object.

Abstract: The present disclosure relates to a method, devices, and storage medium for training neural networks based on memory scores. The said method comprises: establishing the memory scores of a plurality of first-sample images in the library, from their training ages and training indicators, and a preset discount rate; determining a plurality of second-sample images from these memory scores and a preset first count, and using them to establish the first training set; training the neural network by using the first training set, with the said neural network is used for defect detection. The neural network training method in the disclosed embodiment reduces the size of the training set and shortens the time to converge, thereby improving training efficiency.

Abstract: A defect detection device includes an image capturing component for capturing one or more images of an object to be inspected; a motion component configured to grasp or manipulate the object or the image capturing component; and a computing device configured to perform a defect detection method, including determining a plurality of first image capturing poses of the object to be inspected; determining a first defect probability for each particular first image capturing pose; establishing a probability matrix based on the first defect probabilities; subdividing the probability matrix into a plurality of submatrices according to preset dimensions for each submatrix; determining a second defect probability for each of the plurality of submatrices; setting a maximum value of the second defect probabilities as a third defect probability of the object to be inspected; and comparing the third defect probability to a threshold to determine whether the object is defective.