Abstract: A computer system is configured to receive a dataset of image-derived features for a plurality of images, reduce the dimensionality of this dataset, identify clusters within the dimensionally-reduced dataset, and generate a visual representation of the datapoint of the dimensionally-reduced dataset as icons grouped by cluster. User input is received to apply user classification labels to the images for inclusion in a training dataset. A user interface is useable to present information to the user and receive information from the user to facilitate the application of user classification labels.

Abstract: A computer system is configured to receive a dataset of image-derived features for a plurality of images, reduce the dimensionality of this dataset, identify clusters within the dimensionally-reduced dataset, and generate a visual representation of the datapoint of the dimensionally-reduced dataset as icons grouped by cluster. User input is received to apply user classification labels to the images for inclusion in a training dataset. A user interface is useable to present information to the user and receive information from the user to facilitate the application of user classification labels.

Abstract: An automated identification of the types of white blood cells in a blood sample facilitates the manual identification of cancerous or other abnormal blood cells in the sample. Classifiers are predetermined for each type of white blood cell and subsequently used to automatically process images of cells in a sample stained with a nuclear dye or stain. The classifiers each comprise a linear weighted combination of morphometric and photometric features previously selected for white blood cells that were identified using monoclonal antibody stains. Red blood cells and excess fluid are removed from a sample being processed upstream of an imaging region of the imaging system. A plurality of different types of images are produced for each cell by the imaging system enabling automated identification of the white blood cells. Images of any cells not thus identified are manually reviewed to detect cancerous or abnormal cells.

Abstract: A method to perform spatial alignment and spectral compensation for a multi-channel flow imaging system that acquires image data from a single imaging region is disclosed in U.S. Pat. No. 7,079,708. The spatial corrections disclosed therein are static, and do not vary unless the alignment of optical components in the imaging system or the specific detector are modified. However, when image data is acquired from two different imaging regions that are spaced apart along an axis of motion between the object being imaged and the imaging system, dynamic spatial offsets are induced between image data acquired from a first imaging region and image data acquired from a second, spaced apart imaging region. The dynamic spatial offsets are a function of an error in an estimated velocity of the object as it moves between the imaging regions, and may vary from object to object. Techniques for correcting dynamic spatial offsets are disclosed.

Abstract: An automated identification of the types of white blood cells in a blood sample facilitates the manual identification of cancerous or other abnormal blood cells in the sample. Classifiers are predetermined for each type of white blood cell and subsequently used to automatically process images of cells in a sample stained with a nuclear dye or stain. The classifiers each comprise a linear weighted combination of morphometric and photometric features previously selected for white blood cells that were identified using monoclonal antibody stains. Red blood cells and excess fluid are removed from a sample being processed upstream of an imaging region of the imaging system. A plurality of different types of images are produced for each cell by the imaging system enabling automated identification of the white blood cells. Images of any cells not thus identified are manually reviewed to detect cancerous or abnormal cells.

Abstract: Photometric and morphometric features derived from multi-mode imagery of cells in flow are used as a cell analyzer to determine if a marker corresponding to a cancer cell or precancerous cell is present in the population of cells imaged. An imaging system simultaneously acquires a plurality of images for each cell passing through the field of view of the imaging system. Acquiring a plurality of different images (i.e., bright field, dark field, and fluorescent images) facilitates the determination of different morphological and morphometric parameters. Simultaneously acquiring the plurality of images enables relatively large populations of cells to be rapidly imaged, so that relatively small numbers of cancer cells in a large population of cells can be detected. Initially, known cancer cells are imaged to enable a marker to be identified. Then, a sample that may include cancer cells is imaged to determine if the marker is present.

Abstract: Photometric and morphometric features derived from multi-mode imagery of cells in flow are used as a cell analyzer to determine if a marker corresponding to a cancer cell or precancerous cell is present in the population of cells imaged. An imaging system simultaneously acquires a plurality of images for each cell passing through the field of view of the imaging system. Acquiring a plurality of different images (i.e., bright field, dark field, and fluorescent images) facilitates the determination of different morphological and morphometric parameters. Simultaneously acquiring the plurality of images enables relatively large populations of cells to be rapidly imaged, so that relatively small numbers of cancer cells in a large population of cells can be detected. Initially, known cancer cells are imaged to enable a marker to be identified. Then, a sample that may include cancer cells is imaged to determine if the marker is present.

Abstract: Photometric and morphometric features derived from multi-mode imagery of cells in flow are used as a cell analyzer to determine if a marker corresponding to a cancer cell or precancerous cell is present in the population of cells imaged. An imaging system simultaneously acquires a plurality of images for each cell passing through the field of view of the imaging system. Acquiring a plurality of different images (i.e., bright field, dark field, and fluorescent images) facilitates the determination of different morphological and morphometric parameters. Simultaneously acquiring the plurality of images enables relatively large populations of cells to be rapidly imaged, so that relatively small numbers of cancer cells in a large population of cells can be detected. Initially, known cancer cells are imaged to enable a marker to be identified. Then, a sample that may include cancer cells is imaged to determine if the marker is present.

Abstract: Copies of original sound recordings are identified by extracting features from the copy, creating a vector of those features, and comparing that vector against a database of vectors. Identification can be performed for copies of sound recordings that have been subjected to compression and other manipulation such that they are not exact replicas of the original. Computational efficiency permits many hundreds of queries to be serviced at the same time. The vectors may be less than 100 bytes, so that many millions of vectors can be stored on a portable device.

Abstract: A structured morphology filtering method is disclosed for filtering an image for high precision machine vision metrology using specifically-determined structuring elements to precisely preserve the location of specific features in the filtered image. A selected structuring element shape generally exhibits geometric similarity with at least a portion of the feature to be preserved in the filtered image. The structuring element may be oriented to corresponds to the orientation of the feature to be inspected. For example, for a linear feature to be inspected, the optimal structuring element is a line or narrow rectangle at the same orientation, while for images of circles, it is a circle. The orientation of the structuring element may be determined or adjusted automatically during a set of automatic inspection operations, based on an automatic determination of the orientation of the feature to be inspected.

Abstract: Photometric and morphometric features derived from multi-mode imaged. An imaging system simultaneously acquires a plurality of images for each cell passing through the field of view of the imaging system. Acquiring a plurality of different images (i.e., bright field, dark field, and fluorescent images) facilitates the determination of different morphological and morphometric parameters. Simultaneously acquiring the plurality of images enables relatively large populations of cells to be rapidly imaged, so that relatively small numbers of cancer cells in a large population of cells can be detected. Initially, known cancer cells are imaged to enable a marker to be identified. Then, a sample that may include cancer cells is imaged to determine if the marker is present.

Abstract: A method of determining an amount of tilt may include projecting at least two coherent wavefronts toward a target surface, the wavefronts reflecting from the target surface to create an interference fringe pattern on a detector, and transmitting a beam toward the target surface, the transmitted beam reflecting from the target surface to form a beam spot on the detector. A fringe pitch indicative of a distance to the target surface may be determined based on the interference fringe pattern. A displacement on the detector of the beam spot, relative to a nominal location of the beam spot when the target surface is at a nominal angle of incidence relative to the beam, may be determined. The amount of tilt of the target surface relative to the nominal angle of incidence, may be determined based on the displacement of the beam spot and the determined fringe pitch.

Abstract: Copies of original sound recordings are identified by extracting features from the copy, creating a vector of those features, and comparing that vector against a database of vectors. Identification can be performed for copies of sound recordings that have been subjected to compression and other manipulation such that they are not exact replicas of the original. Computational efficiency permits many hundreds of queries to be serviced at the same time. The vectors may be less than 100 bytes, so that many millions of vectors can be stored on a portable device.

Abstract: Improved user interface methods facilitate navigation and programming operations for a magnified machine vision inspection system. Large composite images are determined and stored. The composite images include workpiece features that are distributed beyond the limits of a single magnified field of view of the machine vision system. Despite their size, the composite images may be recalled and displayed in a user-friendly manner that approximates a smooth, real-time, zoom effect. The user interface may include controls that allow a user to easily define a set of workpiece features to be inspected using a composite image, and to easily position the machine vision system to view those workpiece features for the purpose of programming inspection operations for them.

Abstract: Copies of original sound recordings are identified by extracting features from the copy, creating a vector of those features, and comparing that vector against a database of vectors. Identification can be performed for copies of sound recordings that have been subjected to compression and other manipulation such that they are not exact replicas of the original. Computational efficiency permits many hundreds of queries to be serviced at the same time. The vectors may be less than 100 bytes, so that many millions of vectors can be stored on a portable device.

Abstract: A position sensor using a novel structured light generating scale or target member is provided. An imaging array is capable of measuring the relative translation and orientation of the structured light generating scale or target member in X, Y, Z, yaw, pitch, and roll (“6D”) simultaneously, and with high precision. The target member includes an array of lenses that provide an array of structured light patterns that diverge, converge, or both, to change the size of the corresponding structured light image as a function of the “Z” coordinate of the relative position, in various embodiments. The X-Y position of each individual structured light image on the imaging array varies with the relative X-Y position of the structured light generating target member, and the shape of structured light image changes as a function of the relative angular orientation.

Abstract: Improved user interface methods facilitate navigation and programming operations for a magnified machine vision inspection system. Large composite images are determined and stored. The composite images include workpiece features that are distributed beyond the limits of a single magnified field of view of the machine vision system. Despite their size, the composite images may be recalled and displayed in a user-friendly manner that approximates a smooth, real-time, zoom effect. The user interface may include controls that allow a user to easily define a set of workpiece features to be inspected using a composite image, and to easily position the machine vision system to view those workpiece features for the purpose of programming inspection operations for them.

Abstract: A method of determining an amount of tilt may include projecting at least two coherent wavefronts toward a target surface, the wavefronts reflecting from the target surface to create an interference fringe pattern on a detector, and transmitting a beam toward the target surface, the transmitted beam reflecting from the target surface to form a beam spot on the detector. A fringe pitch indicative of a distance to the target surface may be determined based on the interference fringe pattern. A displacement on the detector of the beam spot, relative to a nominal location of the beam spot when the target surface is at a nominal angle of incidence relative to the beam, may be determined. The amount of tilt of the target surface relative to the nominal angle of incidence, may be determined based on the displacement of the beam spot and the determined fringe pitch.

Abstract: A position sensor using a novel optical path array (OPA) element, an angle-selective spatial filter, and an imaging array is capable of measuring the translation and orientation relative to a target member in X, Y, Z, yaw, pitch, and roll (“6D”) simultaneously, and with high precision. A target member includes an array of target points surrounded by a contrasting surface. The position sensor uses the OPA element in combination with the angle-selective spatial filter in a target point imaging arrangement such that the imaging array of the position sensor only receives light rays that enter the OPA element according to an operable cone angle ?. Accordingly, each target point generally produces a ring-shaped image having a size on the imaging array that varies with the Z position of each target point. The X-Y position of each target point image on the imaging array varies with the X-Y position of each target point.

Abstract: A structured morphology filtering method is disclosed for filtering an image for high precision machine vision metrology using specifically-determined structuring elements to precisely preserve the location of specific features in the filtered image. A selected structuring element shape generally exhibits geometric similarity with at least a portion of the feature to be preserved in the filtered image. The structuring element may be oriented to corresponds to the orientation of the feature to be inspected. For example, for a linear feature to be inspected, the optimal structuring element is a line or narrow rectangle at the same orientation, while for images of circles, it is a circle. The orientation of the structuring element may be determined or adjusted automatically during a set of automatic inspection operations, based on an automatic determination of the orientation of the feature to be inspected.