Kabushiki Kaisha Ouyo Keisoku Kenkyusho

Agency of Industrial Science and Technology

Aoi Studio Kabushiki Kaisha

Shigesumi Kuwashima

Nobuyuki Otsu

Takio Kurita

Takahito Suzuki

Chikashi Seki

Kunio Kita

An adaptive learning type general purpose image measurement and recognition method includes the steps of extracting a large number of basic initial features having values which are invariant to parallel displacement of an object to be caught in an image frame and which have additivity with respect to the image frame and performing statistical feature extraction having a learning function on the basis of a multivariate analysis method applied to the extracted initial features to thereby adaptively enable use for various types of measurement.

Method for adaptive learning type general purpose image measurement and

An adaptive learning type general purpose image measurement and recognition method includes the steps of extracting a large number of basic initial features having values which are invariant to parallel displacement of an object to be caught in an image frame and which have additivity with respect to the image frame, and performing statistical feature extraction having a learning function on the basis of a multivariate analysis method applied to the extracted initial features to thereby adaptively enable use for various types of measurement. Further, An adaptive learning type general purpose image measurement and recognition apparatus includes a pick-up device for picking up objects of measurement in two dimensions, an image cutting-out device for partially cutting out video signals of the picked-up images, a correlating device for computing Nth order autocorrelation of the cut-out data, a memory device for storing coefficients of the autocorrelation, and an arithmetic controlling device respetive devices and for carrying out multivariate analysis based on the computed values to thereby adaptively enable use for various types of measurement.

Method and apparatus for adaptive learning type general purpose image

The concept of windows has been widely used in the technical field of image processing. Since the conventional method was necessary to define the shape of the windows so that one window does not include more than one subject, the algorithm of image processing including determination of the windows became inevitably complicated to require complex hardware. Further, the prior method was difficult to process the images at high speed. According to the present invention, there is provided an image processing method which enables efficient processing by giving priority order to windows and processing images in accordance with the predetermined priority order.

Priority order of windows in image processing

A mark (something like a reflecting tape which reflects light in the coaxial direction when the light is applied, or an infrared radiant) attached on an object is caught with the coaxial light (visible radiation, infrared light) by a TV camera or infrared TV camera, and video signals from the camera are processed in binary in an image processing apparatus. Consequently, a high luminous mark appears on a binary image, and the primary moment center values (so called 'centroid') on the high luminous binary image are calculated on real time, for instance, every 1/60 second, and fed back to an electrically-powered driving part of a lighting unit, so that the lighting is always applied to the position. An electrically-powered turn table automatically operates the control of the lighting directions by the signals feed backed. By carrying out a series of controls precisely as mentioned above, the lighting can be applied by the lighting unit tracking automatically the position of the moving mark. The operability of a lighting unit regarded as a serious problem for an operator usually operating many lighting units can be improved to allow an effective lighting operation.

Growing community of inventors

Tokyo, Japan

Shigesumi Kuwashima