U.S. Philips Corporation

Hitachi, Ltd.

Johannes A Bernsen

Seiji Kashioka

Eric H Persoon

Geert Nijholt

Reinier Van Den Boomgaard

During filtering of an image having different regions, the operation of a filter unit is adapted in response to the detection of a boundary between these regions. To this end, one or more intermediate signal values stored in the filter unit for recursive use and derived from signal values from a first region are replaced before any signal value from the second region is filtered. The intermediate signal values are replaced by one or more signal values from the second region, or by filtered signal values derived from the signal values from the second region.

Filter device comprising a recursive filter unit, method of filtering,

Method and apparatus for determining the presence of a predetermined pattern in a spatial field. The invention involves selecting a collection of amplitude values on the basis of location within a multivalued-amplitude field and arranging sequences of amplitude values in order of their magnitude. Amplitude values occurring at predetermined ranks are chosen as characteristic quantities representative of the portion of the field under consideration. By inter-relating a plurality of these characteristic quantities, pattern recognition is accomplished. Preferably rank value filters are employed for producing the characteristic quantities.

Method of processing pattern data on the basis of ranked amplitude

In detecting patterns in data arranged in a spatial field of one, two or higher dimension, a pattern contrast signal is evaluated which measures the similarity to the pattern of the data around a location in the spatial field. The pattern is subsequently detected if the pattern contrast signal is sufficiently strong. There is, however, a risk of erroneous detection when a coincidental noise configuration causes an increased pattern contrast signal. The invention provides for a method and apparatus to discriminate against such erroneous detections. For this purpose, the method observes the extent to which a would-be pattern dominates the effect on the pattern contrast signal of the distribution of data values in the surroundings of the location, this distribution being taken to be typical of noise. In one embodiment, this is done by confirming that the pattern is sufficiently strong to suppress any other would-be pattern detections in the surroundings. In another embodiment, this is done by confirming that the pattern contrast signal responds as expected in the noise free case when the spatial field is rescaled around the location before evaluating the pattern contrast signal.

Method and device for detecting patterns adapted automatically for the

In a detecting patterns in data arranged in a spatial field of one, two or higher dimension, a pattern contrast signal is evaluated which measures the similarity to the pattern of the data around a location in the spatial field. The pattern is subsequently detected if the pattern contrast signal is sufficiently strong. There is, however, a risk of erroneous detection when a coincidental noise configuration causes an increased pattern contrast signal. The invention provides for a method and apparatus to discriminate against such erroneous detections. For this purpose, the method observes the extent to which a would-be pattern dominates the effect on the pattern contrast signal of the distribution of data values in the surroundings of the location, this distribution being taken to be typical of noise. In one embodiment, this is done by confirming that the pattern is sufficiently strong to suppress any other would-be pattern detections in the surroundings. In another embodiment, this is done by confirming that the pattern contrast signal responds as expected in the noise free case when the spatial field is rescaled around the location before evaluating the pattern contrast signal.

A method of recognizing a pattern in a field having a multi-valent amplitude, and a device for performing the method, uses association with a reference mask which is composed of logic high units and logic low units. The reference mask is sub-divided into two sub-masks, units having the same logic value belonging to the same sub-mask. The first and the second sub-mask contain relevant logic high units and logic low units, respectively. After the positioning of the sub-masks in the field, a relatively lowest and a relatively highest value of the amplitude are determined within the window of the first and the second sub-mask, respectively. When the difference formed by the relatively lowest value minus the relatively highest value is not negative, the desired pattern can in principle be recognized in the field. The device for performing the method includes at least one rank-value filter for determining the relatively highest value and the relatively lowest value. There is also determined the sign of the difference between the relatively lowest value and the relatively highest value.

Method of recognizing a pattern in a field having a multi-valent

A pipeline system which is suitable for the processing of data accompanied by data identifying signals comprises interconnected modules which process the data and which controllably transfer the data to one another. Data processing in a module introduces a delay of the data. For correct data processing in a module, the data identifying signals must be in synchronism with the data. In a pipeline system in accordance with the invention, the data identifying signals are applied in parallel to all modules, these signals being delayed in a module by a delay means until the data whereto the signals relate have reached the relevant module. The foregoing has the advantage that the number of connections between the modules is reduced.

Growing community of inventors

Eindhoven, Netherlands

Johannes A Bernsen