Canon Kabushiki Kaisha

Canon Information Systems Research Australia Pty. Ltd.

Delphine Anh Dao Le

Alison Joan Lennon

Mathieu Hitter

The method discloses a method of segmenting an image. The method firstly () allocates pixels as seeds in areas of an image as a function of the luminance of the pixels and the size of those areas. The method then grows () regions from said seeds so as to segment the image into a number of regions. The method considers a number of pixels that border the growing regions and the pixel that is most similar in luminance to a region it borders is appended () to that region. The method then updates () the luminance of the appended region. The method continues until there are no more pixels bordering the growing regions. The method then encodes the segmented image (). It does this by splitting () the image into a number of rectangular sub-images in a quadtree manner until each rectangular sub-image comprises a segmented image forming the dominant portion of the sub-image. The method then merges () rectangular sub-images which have a common dominant portion and share a common edge.

Method and apparatus for segmenting images

A method of interpreting an image using a statistical or probabilistic interpretation model is disclosed. The image has associated therewith contextual information. The method comprises the following steps: providing the contextual information associated with the image for analysis; analyzing the additional contextual information to identify predetermined features relating to the image; and biasing the statistical or probabilistic interpretation model in accordance with the identified features.

Image interpretation method and apparatus

The method discloses a method of segmenting moving objects and determining their motion from video data. The method generates ( ) a motion vector field by comparing the current frame of video data with the previous frame, where a motion vector is determined for each pixel of the current frame. The method then generates ( ) a motion parameter field by sliding a window over the motion vector field of the current frame, and determining a set of motion parameters for each pixel based on a motion model of the motion vector field. The method then distributes ( ) seeds throughout the motion parameter field so that fewer seeds are allocated to those areas of the motion parameter field having homogeneous motion parameter sets. The method then grows ( ) regions from said seeds so as to segment the motion parameter field into a number of regions. The method considers a number of pixels that border the growing regions and the pixel that has a motion vector that is most similar to the motion parameter field of a region it borders is appended to that region. The method then updates the set of motion parameters of the appended region. The method continues until there are no more pixels bordering the growing regions.

Segmenting moving objects and determining their motion

A method of smoothing jagged edges in graphical data. The method detects one or more edge of a selected pixel in the graphical data, dependent upon intensities of the selected pixel and another pixel surrounding a respective site of one or more edges and predetermied gradients of at least the selected pixel and the respective site surrounding the pixel. An adapted convolution mask is applied to the selected pixel and a predetermined neighborhood of pixels containing the selected pixel. The coefficient values of the convolution mask are dependent upon one or more detected edges.

Method for smoothing jagged edges in digital images

Digital image signal interpretation is the process of understanding the content of an image through the identification of significant objects or regions in the image and analysing their spatial arrangement. Traditionally the task of image interpretation required human analysis. This is expensive and time consuming, consequently considerable research has been directed towards constructing automated image interpretation systems. A method of interpreting a digital video signal is disclosed whereby the digital video signal has contextual data. The method comprising the steps of firstly, segmenting the digital video signal into one or more video segments, each segment having a corresponding portion of the contextual data. Secondly, analysing each video segment to provide a graph at one or more temporal instances in the respective video segment dependent upon the corresponding portion of the contextual data.

Growing community of inventors

Drummoyne, Australia

Delphine Anh Dao Le