Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.08289 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914461480124416 |
|---|---|
| author | Kotov, Matvei Ciccarelli, Lorenzo |
| author_facet | Kotov, Matvei Ciccarelli, Lorenzo |
| contents | In this paper, we consider an image coding process consisting of the following four steps: a direct transformation, a direct quantization, an inverse quantization, and an inverse transformation, where Hadamard transforms are used for the transformation steps and a dead-zone quantizer is used for the quantization. The aim of this paper is to provide a theoretical tool for analyzing this process. We discuss error bounds for this process and bounds on the largest absolute value that the components of the result can attain. In order to obtain these bounds, we use methods of linear algebra and properties of Hadamard matrices. The obtained formulae depend on the size of the matrices, the parameters of the quantizer and the dequantizer, and a bound on the source values. Knowing the error bounds helps control the trade-off between compression efficiency and output quality. Knowing the bounds on the largest absolute value helps decide how many bits are needed to store the result. In addition, we demonstrate a connection between the norm $\|\mathbf{H}\|_{\infty, 1}$ of a Hadamard matrix $\mathbf{H}$ and the maximal excess $σ([\mathbf{H}])$ of the equivalence class containing $\mathbf{H}$. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08289 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Error analysis of quantization combined with Hadamard transforms Kotov, Matvei Ciccarelli, Lorenzo Combinatorics 68U10, 15B34 In this paper, we consider an image coding process consisting of the following four steps: a direct transformation, a direct quantization, an inverse quantization, and an inverse transformation, where Hadamard transforms are used for the transformation steps and a dead-zone quantizer is used for the quantization. The aim of this paper is to provide a theoretical tool for analyzing this process. We discuss error bounds for this process and bounds on the largest absolute value that the components of the result can attain. In order to obtain these bounds, we use methods of linear algebra and properties of Hadamard matrices. The obtained formulae depend on the size of the matrices, the parameters of the quantizer and the dequantizer, and a bound on the source values. Knowing the error bounds helps control the trade-off between compression efficiency and output quality. Knowing the bounds on the largest absolute value helps decide how many bits are needed to store the result. In addition, we demonstrate a connection between the norm $\|\mathbf{H}\|_{\infty, 1}$ of a Hadamard matrix $\mathbf{H}$ and the maximal excess $σ([\mathbf{H}])$ of the equivalence class containing $\mathbf{H}$. |
| title | Error analysis of quantization combined with Hadamard transforms |
| topic | Combinatorics 68U10, 15B34 |
| url | https://arxiv.org/abs/2604.08289 |