Canon Kabushiki Kaisha

Commonwealth Scientific and Industrial Research Organisation

Timothy John Wark

James Philip Andrew

James Austin Besley

George Politis

Yu-Ling Chen

Eric Wai-Shing Chong

Steven Richard Irrgang

Dmitri Katchalov

Michael Jan Lawther

Raja Jurdak

Reuben Kan

Muhammad Majid

This invention concerns remote sensor networks, and particularly energy management for wireless sensor networks. In a first aspect the invention is a wireless sensor node specified to operate for a given lifetime, including an onboard computer system and a set of one or more associated sensors. The computer system operates to periodically sample data from each sensor of the set of associated sensors, and to store a multi-state model representing one or more phenomena described by the collected data. And, the computer system operates to calculate a value associated with movement of the phenomena between the states of the multi-state model, and to adjust the rate of sampling of one or more of the set of associated sensors depending on the calculated value. In other aspects the invention is a network of sensor nodes and a method of operation.

Energy management for wireless sensor networks

A method () and system () for classifying a audio signal are described. The method () operates by first receiving a sequence of audio frame feature data, each of the frame feature data characterising an audio frame along the audio segment. In response to receipt of each of the audio frame feature data, statistical data characterising the audio segment is updated with the received frame feature data. The received frame feature data is then discarded. A preliminary classification for the audio segment may be determined from the statistical data. Upon receipt of a notification of an end boundary of the audio segment, the audio segment is classified () based on the statistical data.

Effective audio segmentation and classification

A method (), an apparatus, and a computer program product for automatically producing a compact representation of a colour document are disclosed. In the method, a digital image of a colour-document page is segmented () into connected components in one-pass, block raster order. The digital image of the page is partitioned into foreground and background images using layout analysis () based on compact, connected-component statistics of the whole page. At least one portion of the background image where at least one portion of the foreground image obscures the background image is inpainting () in one-pass block raster order. The foreground and background images are combined () to form a compact document. A method, an apparatus, and a computer program product for segmenting a digital image comprising a plurality of pixels are also disclosed.

Segmenting digital image and producing compact representation

Methods, computer program products and apparatuses are disclosed for processing images to detect and/or remove halftone. In one aspect, a portion of an image comprising a plurality of color channels is selected for processing (), a frequency domain representation of the selected portion is generated for each color channel (), data relating to predetermined regions in the frequency domain representations is processed (), and halftone present in the selected portion is detected based on a result of the processing the data (). In another aspect, an image is arranged into a plurality of bands each comprising a predetermined number of consecutive lines of pixels, each of the plurality of bands is arranged into a plurality of tiles of pixels, one of the plurality of tiles is selected as a selected portion, data relating to the selected portion is processed and halftone present in the selected portion is detected based on a result of the processing and on a result of processing at least one other portion of the image. Halftone may be detected based on spatial characteristics or frequency characteristics within the selected portion and detected halftone may be removed from spatial domain data or frequency domain data relating to the selected portion.

Halftone detection and removal

A method () and apparatus () for classifying a homogeneous audio segment are disclosed. The homogeneous audio comprises a sequence of audio samples (x(n)). The method () starts by forming a sequence of frames (-) along the sequence of audio samples (x(n)), each frame (-) comprising a plurality of the audio samples (x(n)). The homogeneous audio segment is next divided () into a plurality of audio clips (-), with each audio clip being associated with a plurality of the frames (-). The method () then extracts () at least one frame feature for each clip (-). A clip feature vector (f) is next extracted from frame features of frames associated with the audio clip (-). Finally the segment is classified based on a continuous function during the distribution of the clip feature vectors (f).

Robust detection and classification of objects in audio using limited training data

A method () and apparatus () for segmenting a sequence of audio samples into homogeneous segments (and) are disclosed. The method () forms a sequence of frames (to) along the sequence of audio samples, and extracts, for each frame, a data feature. The data features form a sequence of data features. Transition points in the sequence of data features are thin detected by applying the Bayesian Information Criterion to the sequence of data features. The transition points define the homogeneous segments (and). Preferably the data feature is single-dimensional and a leptokurtic distribution is used as an event model in the Bayesian Information Criterion.

Growing community of inventors

The Gap, Australia

Timothy John Wark