Aritmetic Compression of Weighted Finite Automata
نویسندگان
چکیده
Karel Culik and the first author have demonstrated how Weighted Finite Automata (WFA) provide a strong tool for image compression [1, 2, 3, 4]. In the present article we introduce an improved method for the last step of the compression algorithm: for compressing the WFA that approximates the given image. Our method is based on arithmetic compression of sparse matrices.
منابع مشابه
On Succinct Representations of Textured Surfaces by Weighted Finite Automata
Generalized finite automata with weights for states and transitions have been successfully applied to image generation for more than a decade now. Bilevel images (black and white), grayscaleor color-images and even video sequences can be effectively coded as weighted finite automata. Since each state represents a subimage within those automata the weighted transitions can exploit self-similarit...
متن کاملHyper-minimisation of deterministic weighted finite automata over semifields
Hyper-minimisation of deterministic finite automata is a recently introduced state reduction technique that allows a finite change in the recognised language. A generalisation of this lossy compression method to the weighted setting over semifields is presented, which allows the recognised formal power series to differ for finitely many input strings. First, the structure of hyper-minimal deter...
متن کاملBetter Compression Performance by Compressing Sliced Bit Planes of Image Using Weighted Finite Automata
Weighted Finite Automata (WFA) is an efficient structure for digital image storage and is successfully applied for the image compression purpose. This paper introduces the idea of slicing image into eight bit planes and then compressing each bit planes individually using proposed WFA inference algorithm. Furthermore, the proposed algorithm uses the concept of invertible resolution wise mapping ...
متن کاملM . Droste and P . Gastin Weighted automata and weighted logics Research Report LSV - 05 - 13 July 2005
Weighted automata are used to describe quantitative properties in various areas such as probabilistic systems, image compression, speech-to-text processing. The behaviour of such an automaton is a mapping, called a formal power series, assigning to each word a weight in some semiring. We generalize Büchi’s and Elgot’s fundamental theorems to this quantitative setting. We introduce a weighted ve...
متن کاملPattern - Matching Problems for
The power of weighted nite automata to describe very complex images was widely studied, see [5, 6, 7]. Finite automata can be also used as an e ective tool for compression of twodimensional images. There are some software packages using this type of compression, see [12, 6]. We consider the complexity of some pattern-matching problems for two-dimensional images which are highly compressed using...
متن کامل