Compression for visual pattern recognition

Abstract

To date, computer science solves pattern recognition problems by highly task specific algorithms. Searching for a generic, unifying principle of pattern recognition, Benedetto et al. showed that compression is a good candidate: For the domain of text, approximating the mutual information of patterns by the achievable compression factors allows to detect similarity with surprising accuracy. Here we show that this principle is much more general than expected, since the common compressor gzip is able to solve a major computer vision problem, the classification of image categories. This result is remarkable for three reasons: Firstly, the applied compression programs were never designed for pattern recognition tasks of any kind; secondly, hardly any other method is able to deal with patterns as different as text and pixel images alike without any modification; thirdly, compression solves the entire task in a single step, without any data preprocessing, feature extraction or the need for parametrization. We will discuss the theoretical background of this finding and point out the role of compression in a yet to develop general theory of pattern recognition.

The Cyborg Astrobiologist: matching of prior textures by image compression for geological mapping and novelty detection

“…The juxtaposition in our implementation is left–right, but we have not compared our implementation to an up–down juxtaposition. Various image-compression programmes were investigated by Heidemann & Ritter (2008a, b), and their optimal image compressor for texture classification was gzip , which we have chosen as our image compressor 1 . Gzip relies on Huffman entropy coding (Huffman 1952; see also Saravanan & Ponalagusamy 2010) and the Lempel–Ziv algorithm (Lempel & Ziv 1977; see also Kärkkäinen et al .…”

“…The Lempel-Ziv algorithm (LZ77) eliminates duplicate strings of symbols using 'sliding-window compression' (referring to the buffer window that records the previously-observed symbols in the data stream) 3 . Quoting Heidemann and Ritter (2008b): "The Lempel-Ziv algorithm (LZ77) (Lempel and Ziv, 1977), which detects repeatedly occurring symbol sequences within the data, such that a dictionary can be established. A repeated symbol sequence can then be replaced by the symbol defined in the dictionary."…”

“…A repeated symbol sequence can then be replaced by the symbol defined in the dictionary." 4 The gzip algorithm outperforms (Heidemann and Ritter, 2008b) the correlation technique and the histogram-based matching technique for the texture-classification, object-recognition and image-retrieval tasks.…”

“…Herein, using data acquired by the Astrobiology Phone-cam at a geological field site (Fig. 1), we test an image-comparison technique of Heidemann and Ritter (2008a, b), which uses image compression and is capable of (i) detecting novel (coloured) textures in a series of images as well as of (ii) alerting the user to the similarity of a new image to a previously observed texture. We first implemented this technique in 2010 (Bonnici et al 2010), but did not test this software at a geological field site until now.…”

“…This is one shortcoming of this dictionary-based technique for data compression. In the future, as suggested by Heidemann and Ritter (2008a,b), other data-compression techniques could be used (or developed) that can handle better the inherent variability of RGB pixel values. Such techniques would be in principle lossy, though lossy compression in the JPEG implementation does not work well, due to the discrete cosine transform that was utilized by JPEG (Heidemann and Ritter, 2008a).…”