Generating a textual description of an image includes classifying an image represented by image data into a domain-specific category, and segmenting one or more elements in the image data based on the domain-specific category. Each element of the one or more elements is compared to a domain-independent model to detect one or more statistical anomalies in the one or more elements. The one or more detected statistical anomalies are characterized using one or more domain-independent text phrases. The one or more domain-independent text phrases are converted to one or more domain-specific descriptions based upon the domain-specific category.

A method and system for automatic location of a target treatment structure, such as a pulmonary vein ostium, from an anatomical image. The method includes calculating a most likely path of blood flow through a pulmonary vein based on a cross-sectional area minimization technique and calculating pulmonary vein geometry as a function of length. For example, a pulmonary vein ostium may be located by analyzing a change in pulmonary vein dimensional size or other anatomical factors, such as absolute size. The method may also include determining tissue thickness at the pulmonary vein ostium or other treatment size for treatment dose optimization. The method may be an algorithm performed by a processing unit of a navigation system or other component of a medical system.

The present disclosure provides a method and device for counting pedestrians. The method comprises: reading a pedestrian depth image, and comparing the pedestrian depth image and a pre-acquired environmental mean image to acquire a foreground image; dividing the foreground image into a plurality of regions, detecting whether or not there is a step in an edge pixel point of each region, and detecting whether or not the region surface formed in each region coincides with the curved surface of the head top; determining that the currently detected region is the region of the head top when there is a step in the point and when the region surface coincides with the curved surface; and counting and outputting the number of pedestrians according to the region of the head top determined from the pedestrian depth image and the region of the head top determined from an adjacent pedestrian depth image.

A first detection unit detects a breast region and a skin line from a breast image, and a first index value calculation unit calculates a first index value indicating the single composition degree of the breast region. A second detection unit detects a boundary between the adipose tissue and the mammary gland tissue in a predetermined range from the skin line toward the inside of the breast region in the breast image. A second index value acquisition unit acquires a second index value indicating the degree of clogging of mammary glands with respect to the breast region based on at least one of the strength of the boundary or the distance from the skin line. An identification unit identifies the type of the breast based on the first and second index values.

An image data processing method and a printing system for printing technology are provided. The image includes a first bitmap image. The image data processing method includes: dividing the first bitmap image into a plurality of regions, selecting sampling positions in each of the plurality of regions, performing sampling to acquire sample points, and rearranging the sample points to form a second bitmap image. The second bitmap image is different from the first bitmap image.

A method and system for automatic location of a target treatment structure, such as a pulmonary vein ostium, from an anatomical image. The method includes calculating a most likely path of blood flow through a pulmonary vein based on a cross-sectional area minimization technique and calculating pulmonary vein geometry as a function of length. For example, a pulmonary vein ostium may be located by analyzing a change in pulmonary vein dimensional size or other anatomical factors, such as absolute size. The method may also include determining tissue thickness at the pulmonary vein ostium or other treatment size for treatment dose optimization. The method may be an algorithm performed by a processing unit of a navigation system or other component of a medical system.

An imaging system (100) includes a sub-resolution luminal narrowing detector (112) which detects sub-resolution narrowing of a vessel lumen in an image volume by a centerline profile analysis and computes a sub-resolution determined diameter by modifying an approximated visible lumen diameter with the detected sub-resolution narrowing.

A circular scratch inspection apparatus includes: a camera capturing an image of a workpiece surface around a hole; illumination device emitting light to the workpiece surface around the hole, the light being reflected on the workpiece surface is not directly incident on the camera; and image processor. The image processor: generates a second-derivative image by performing secondary differentiation on luminance values in an actual image obtained by the camera; generates a second-derivative curve for each of a plurality of ruler lines, extending radially from the hole center and are set in an inspection target region on the workpiece surface; counts a first reference number of times for each ruler line; calculates a first reference total number of times; and determines presence or absence of a circular scratch by using the first reference total number of times.

In part, the invention relates to systems and methods of calibrating a plurality of frames generated with respect to a blood vessel as a result of a pullback of an intravascular imaging probe being pullback through the vessel. A calibration feature disposed in the frames that changes between a subset of the frames can be used to perform calibration. Calibration can be performed post-pullback. Various filters and image processing techniques can be used to identify one or more feature in the frames including, without limitation, a calibration feature, a guidewire, a side branch, a stent strut, a lumen of the blood vessel, and other features. The feature can be displayed using a graphic user interface.

Generating a textual description of an image includes classifying an image represented by image data into a domain-specific category, and segmenting one or more elements in the image data based on the domain-specific category. Each element of the one or more elements is compared to a domain-independent model to detect one or more statistical anomalies in the one or more elements. The one or more detected statistical anomalies are characterized using one or more domain-independent text phrases. The one or more domain-independent text phrases are converted to one or more domain-specific descriptions based upon the domain-specific category.