Several color object irecognition methods that are based on image retrieva1 algorithms attempt to discount changes of illuminlztion in order to increase performance when test image illumination conditions differ from those that obtained when the image database was created. Here we extend the seminal method of Swain and Ballard to discount changing illumination. The new method is based on the first stage of the simplest color indexing medhod, which uses angular invariants between color image and edge image channels. That method Jirst normalizes image channels, and then effectively discards much of the remaining information. Here we adopt the color-normalization stage as an adequate color constancy step. Further, we replace 30 color histograms by 20 chromaticity histograms. Treating these as images, we implement the method in a compressed histogram-image domain using a combination of wavelet compression and Discrete Cosine Transform (DCT to fully exploit the technique of low-pass filtering / or eficiency. Results are very.encouraging, wath substantially better performance than other methods tested. The method is also fast, in that the indexing process is entirely carried out in the compressed domain and uses a feature vector of only 36 or 72 values.

Generalised color moments combine shape and color information and put them on an equal footing. Rational expressions of such moments can be designed, that are invariant under both geometric deformations and photometric changes. These generalised color moment invariants are e#ective features for recognition under changing viewpoint and illumination. The paper gives a systematic overview of such moment invariants for several combinations of deformations and photometric changes. Their validity and potential is corroborated through a series of experiments. Both the cases of indoor and outdoor images are considered, as illumination changes tend to di#er between these circumstances. Although the generalised color moment invariants are extracted from planar surface patches, it is argued that invariant neighbourhoods o#er a concept through which they can also be used to deal with 3D objects and scenes.

Images or videos may be imaged under different illuminants than models in an image or video proxy database. Changing illumination color in particular may confound recognition algorithms based on color histograms or video segmentation routines based on these. Here we show that a very simple method of discounting illumination changes is adequate for both image retrieval and video segmentation tasks. We develop a feature vector of only 36 values that can also be em used for both these objectives as well as for retrieval of video proxy images from a database. The new image metric is based on a color-channel-normalization step, followed by reduction of dimensionality by going to a chromaticity space. Treating chromaticity histograms as images, we perform an effective low-pass filtering of the histogram by first reducing its resolution via a wavelet-based compression and then by a DCT transformation followed by zonal coding. We show that the color constancy step -- color band normalization -- can...

Many methods for video segmentation rely upon the setting and tuning of thresholds for classifying interframe distances under various difference measures. An approach that has been used with some success has been to establish statistical measures for each new video and identify camera cuts as difference values far from the mean. For this type of strategy the mean and dispersion for some interframe distance measure must be calculated for each new video as a whole. Here we eliminate this statistical characterization step and at the same time allow for segmentation of streaming video by introducing a preprocessing step for illumination--invariance that concomitantly reduces input values to a uniform scale. The preprocessing step provides a solution to the problem that simple changes of illumination in a scene, such as an actor emerging from a shadow, can trigger a false positive transition, no matter whether intensity alone or chrominance is used in a distance measure. Our means of discou...

This thesis describes research into an area of content based image retrieval (CBIR), that of feature indexing for the purpose of rapid retrieval. The techniques in this thesis draw from the field of text IR and demonstrate that individual image extraction algorithms can be optimised for use with an inverted index, which could lead to CBIR systems capable of sub-second retrieval times on collections of millions of images. A novel global feature algorithm...