An image processing apparatus acquires a detection area in an image coordinate system expressing an area of a target object acquired from within an image, and derives a target spatial area in which the detection area is transformed to a corresponding position in a spatial coordinate system that simulates actual space in which the target object is present. In addition, the image processing apparatus identifies a reference physical model that simulates characteristics related to behavior and shape of the target object in the spatial coordinate system. Next, the image processing apparatus compares the target spatial area and the reference physical model on the spatial coordinate system, and corrects position and shape of the target spatial area based on the comparison result. Then, the image processing apparatus transforms the corrected target spatial area to a corresponding position in the image coordinate system, and outputs corrected area information expressing the corrected area.

There is provided an image processing device. A feature point acquiring unit acquires a plurality of feature points of objects of each frame image. A grouping unit performs a grouping process of dividing the feature points into groups corresponding to the individual objects. A masking unit performs a masking process of excluding feature points included in some areas of the frame image. A highlighting-image generating unit generates highlighting images for highlighting objects of the frame image corresponding to the feature point groups. A superimposed-image generating unit superimposes the highlighting images on the frame image, thereby generating a superimposed image.

Systems and methods for detecting anomalies in aviation data communication systems (e.g., air traffic control surveillance systems), include a processor receiving device status information. A variational autoencoder receives and optimizes the device status information and determines whether it qualifies as an anomaly. Optimized device status information is compared to either non-anomalous or anomalous device status data in a latent space of the variational autoencoder. The latent space preferably includes an n-D point scatter plot and hidden vector values. The processor optimizes the device status information by generating a plurality of probabilistic models of the device status information and determining which of the plurality of models is optimal. A game theoretic optimization is applied to the plurality of models, and the best model is used to generate the n-D point scatter plot in latent space. An image gradient sobel edge detector preprocesses the device status information prior to optimization.

An image processing apparatus acquires a detection area in an image coordinate system expressing an area of a target object acquired from within an image, and derives a target spatial area in which the detection area is transformed to a corresponding position in a spatial coordinate system that simulates actual space in which the target object is present. In addition, the image processing apparatus identifies a reference physical model that simulates characteristics related to behavior and shape of the target object in the spatial coordinate system. Next, the image processing apparatus compares the target spatial area and the reference physical model on the spatial coordinate system, and corrects position and shape of the target spatial area based on the comparison result. Then, the image processing apparatus transforms the corrected target spatial area to a corresponding position in the image coordinate system, and outputs corrected area information expressing the corrected area.

Systems and methods for detecting anomalies in aviation data communication systems (e.g., air traffic control surveillance systems), include a processor receiving device status information. A variational autoencoder receives and optimizes the device status information and determines whether it qualifies as an anomaly. Optimized device status information is compared to either non-anomalous or anomalous device status data in a latent space of the variational autoencoder. The latent space preferably includes an n-D point scatter plot and hidden vector values. The processor optimizes the device status information by generating a plurality of probabilistic models of the device status information and determining which of the plurality of models is optimal. A game theoretic optimization is applied to the plurality of models, and the best model is used to generate the n-D point scatter plot in latent space. An image gradient sobel edge detector preprocesses the device status information prior to optimization.

There is provided an image processing device. A feature point acquiring unit acquires a plurality of feature points of objects of each frame image. A grouping unit performs a grouping process of dividing the feature points into groups corresponding to the individual objects. A masking unit performs a masking process of excluding feature points included in some areas of the frame image. A highlighting-image generating unit generates highlighting images for highlighting objects of the frame image corresponding to the feature point groups. A superimposed-image generating unit superimposes the highlighting images on the frame image, thereby generating a superimposed image.

A method for computer recognition of a shape described by a path. The method includes: obtaining a parameterized version of the path having a plurality of points; calculating a plurality of tangent angles for the plurality of points; determining a distribution for the plurality of tangent angles; obtaining a plurality of reference distributions for a plurality of reference shapes; comparing the distribution with the plurality of reference distributions; and matching, based on the comparing, the shape to one of the plurality of reference shapes.

A handwriting geometry recognition and calibration system by using neural network and mathematical feature includes: a pre-processor for pre-processing coordinate points of geometric figures from user's handwriting so as to get a plurality of sample points which expresses the geometric figures to be recognized; a neural network connected to the pre-processor for receiving the sample points of the geometric figure and recognizing the geometric figure so as to acquire a coarse class of the geometric figure; and an mathematical logic unit connected to the neural network for receiving recognition results from the neural network, including coarse classifications which are used in a secondary classification by using conventional mathematical recognition logics so as to determine an exact geometry shape of the geometric figure; then the geometric figure being calibrated so as to get a geometry with a regular shape.

A method for cargo counting, a computer equipment, and a storage medium are provided in the disclosure. The method includes the following. Three-dimensional (3D) point cloud data of a set of cargoes within a preset placement region is obtained based on a cargo-counting instruction. Whether the set of cargoes are in a first placement state is determined according to the 3D point cloud data. Based on a determination that the set of cargoes are in the first placement state, a quantity of the set of cargoes is calculated.

According to the present invention, an image recognition system calculates importance of a feature for each target shape recognized in an image and for each type of feature, and determines correctness of a recognition result by comparing the importance with a statistic for each type of feature, for each target shape.