Abstract: Various embodiments provide computer vision techniques for ensuring system- and component-level integrity via component characterization. Various embodiments may comprise computer vision systems that are configured to extract pin information from surface-mount device (“SMD”) contours. Accordingly, the disclosed embodiments may include a computer vision system configured to detect the extremities of each pin in a SMD contour image based on a combination of at least one of machine learning and one or more PCB image collection datasets.

Abstract: Various embodiments of the present disclosure provide systems and methods for classifying and evaluating an electronic object based at least in part on terahertz (THz) data. THz data may comprise time-domain THz data collected via THz time domain spectroscopy of the electronic object. THz data may further comprise frequency-domain THz data, which may be generated based at least in part on the time-domain THz data. A unique THz fingerprint may be generated for the electronic object based at least in part on the THz data. This unique THz fingerprint may be compared to an earlier generated THz fingerprint of the same electronic object to evaluate reliability and consistency. The unique THz fingerprint may also be compared to THz fingerprints or THz data of other electronic objects of the same object type, class, design, and/or the like, to validate the electronic object (e.g., determine if the electronic object is counterfeit).

Abstract: There is a need for more effective and efficient printed circuit board (PCB) design. This need can be addressed by, for example, solutions for performing automated PCB component estimation. In one example, a method includes identifying a plurality of initial component estimations for the PCB; performing a shadow detection segmentation using the plurality of initial component estimations, a non-direct-lighting image, and one or more direct-lighting images to generate a first set of detected PCB components; performing a super-pixel segmentation using the plurality of initial component estimations and the non-direct-lighting-image to generate a second set of detected PCB components; and generating a bill of materials for the PCB based at least in part on the first set of detected PCB components and the second set of detected PCB components.

Abstract: There is a need for more effective and efficient printed circuit board (PCB) design. This need can be addressed by, for example, solutions for performing automated PCB component estimation. In one example, a method includes identifying a plurality of initial component estimations for the PCB; performing a shadow detection segmentation using the plurality of initial component estimations, a non-direct-lighting image, and one or more direct-lighting images to generate a first set of detected PCB components; performing a super-pixel segmentation using the plurality of initial component estimations and the non-direct-lighting-image to generate a second set of detected PCB components; and generating a bill of materials for the PCB based at least in part on the first set of detected PCB components and the second set of detected PCB components.