NEURAL NETWORK FOR BULK SORTING
    1.
    发明申请

    公开(公告)号:US20230011383A1

    公开(公告)日:2023-01-12

    申请号:US17773319

    申请日:2020-11-04

    Abstract: A bulk sorting system for sorting objects in bulk is provided. The bulk sorting system includes: at least one radiation source arranged to radiate the objects, at least one optical sensor arranged to capture reflected radiation of the objects and acquire the reflected radiation as multi- or hyperspectral data; a processing circuit configured to analyze the reflected radiation of the objects by inputting the multi- or hyperspectral data into a convolutional neural network (CNN) with at least two convolutional layers in order to either detect and classify the objects in the multi- or hyperspectral data and/or semantically segment the multi- or hyperspectral data; and a mechanical sorter configured to sort the objects according to their classification and/or segmentation using the analysis of the processing circuit such that different overlapping and/or stacked objects are separated or treated as a single group of objects.

    MATERIAL IDENTIFICATION APPARATUS AND METHOD

    公开(公告)号:US20250146937A1

    公开(公告)日:2025-05-08

    申请号:US18715204

    申请日:2022-12-06

    Inventor: Dirk BALTHASAR

    Abstract: The present invention relates to an apparatus for classification of matter comprising: a scanning element is configured to redirect the at least one illumination beam and to shift a plurality of inspection zones and an irradiated area relative said matter in the first direction. A processing circuitry configured to execute: a second zone collection function configured to collect second zone data based which pertains to said optical radiation emitted by said matter in the second inspection zone, a third zone collection function configured to collect third zone data based which pertains to optical radiation emitted by said matter in the third inspection zone, a classification function configured to classify said matter based on the second zone data and the third zone data.

    METHOD AND SYSTEM FOR TRAINING A NEURAL NETWORK-IMPLEMENTED SENSOR SYSTEM TO CLASSIFY OBJECTS IN A BULK FLOW

    公开(公告)号:US20230206606A1

    公开(公告)日:2023-06-29

    申请号:US18000698

    申请日:2021-06-02

    Inventor: Dirk BALTHASAR

    Abstract: A method of training a neural network stored on a computer-readable storage medium to classify objects in a bulk flow, the method including: providing input image data depicting objects to be classified, which input image data is captured by means of an input imaging sensor of a first sensor technology design; providing auxiliary image data, which auxiliary image data is captured by means of an auxiliary imaging sensor of a second sensor technology design, and which auxiliary image data depicts said or similar objects which are classified in accordance with a predetermined classifying scheme; by means of a processing unit, train the neural network stored on the computer-readable storage medium to classify the depicted objects in the input image data based on classifications of depicted objects in the auxiliary image data, wherein the depicted objects in the input image data correspond to objects in a bulk flow.

    APPARATUS FOR ILLUMINATING MATTER

    公开(公告)号:US20250018433A1

    公开(公告)日:2025-01-16

    申请号:US18715364

    申请日:2022-12-06

    Inventor: Dirk BALTHASAR

    Abstract: We provide an apparatus for detecting matter comprising: An irradiation arrangement adapted to emit one of a first and a second set of illumination beams towards a detection zone where the matter is provided. A detection system receives and analyzes light reflected, emitted and/or scattered by matter in the detection zone. A switching device comprising a carrier movable between a first position, wherein the only the first illumination device in an active illumination position, and a second position wherein only said second illumination device in said active illumination position, and wherein said irradiation arrangement is configured to emit an illumination beam only from the one of said first and second illumination devices arranged in said active illumination position and an actuator for physically moving said carrier from said first position to said second position based on the condition of said first illumination device and/or in response to user-initiated input.

    APPARATUS FOR DETECTING MATTER
    5.
    发明公开

    公开(公告)号:US20230213443A1

    公开(公告)日:2023-07-06

    申请号:US18000549

    申请日:2021-05-04

    CPC classification number: G01N21/39 G01J3/2823 G01N2021/845

    Abstract: An apparatus for detecting matter including: a light source arrangement adapted to emit a first and a second set of light beams towards a first detection zone through which the matter is provided. A spectroscopy system adapted to receive and analyse light which is reflected and/or scattered by matter in the first detection zone. A laser triangulation system including, a laser arrangement adapted to emit a line of laser light towards a second detection zone. A camera-based sensor arrangement configured to receive and analyse light which is reflected and/or scattered by matter in the second detection zone. The received light of the spectroscopy system completely or partially intersects the received light of the camera-based sensor arrangement and/or the line of laser light.

    CLASSIFICATION METHOD AND APPARATUS
    6.
    发明申请

    公开(公告)号:US20200331034A1

    公开(公告)日:2020-10-22

    申请号:US16643244

    申请日:2018-09-03

    Abstract: A method for object classification, including: guiding a continuous stream of objects from a transport mechanism directly into an unsupported path, along which the object stream is fed through a detection region; illuminating the detection region with a radiation band in a first direction; optically scanning the detection region to detect electromagnetic radiation reflected by the at least one object in the detection region by viewing the detection region along a second direction, wherein the directions have a cross-section in the detection region and form an angle relative each other of 10°-80°; analyzing the information; and classifying objects in at least two different directions based on the analysis of the transparency information from the optical scanning; wherein the step of analyzing the information from the optical scanning includes analyzing the objects based on the intensity, spread and texture of the detected radiation from the optical scanning.

Patent Agency Ranking