# Introducing Hann windows for reducing edge-effects in patch-based image segmentation

arXiv: https://arxiv.org/abs/1910.07831 · HTML: https://mlsanity.com/paper/1910.07831/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Oct 2019.

## Built on by (most cited first, 17 of 17)

- [DiffInfinite: Large Mask-Image Synthesis via Parallel Random Patch Diffusion in Histopathology](https://mlsanity.com/paper/2306.13384/) (2023)
- [OMGSR: You Only Need One Mid-timestep Guidance for Real-World Image Super-Resolution](https://mlsanity.com/paper/2508.08227/) (2025)
- [Simulating analogue film damage to analyse and improve artefact restoration on high-resolution scans](https://mlsanity.com/paper/2302.10004/) (2023)
- [Time Series Forecastability Measures](https://mlsanity.com/paper/2507.13556/) (2025)
- [Synthesizing Accurate and Realistic T1-weighted Contrast-Enhanced MR Images using Posterior-Mean Rectified Flow](https://mlsanity.com/paper/2508.12640/) (2025)
- [Localized PCA-Net Neural Operators for Scalable Solution Reconstruction of Elliptic PDEs](https://mlsanity.com/paper/2509.18110/) (2025)
- [AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion](https://mlsanity.com/paper/2605.26130/) (2026)
- [Convolutional Neural Network (CNN) to reduce construction loss in JPEG compression caused by Discrete Fourier Transform (DFT)](https://mlsanity.com/paper/2209.03475/) (2022)
- [Modified watershed approach for segmentation of complex optical coherence tomographic images](https://mlsanity.com/paper/2303.16609/) (2023)
- [ARTeFACT: Benchmarking Segmentation Models on Diverse Analogue Media Damage](https://mlsanity.com/paper/2412.04580/) (2024)
- [Overcoming Labelled Data Scarcity for Defect Classification in Scanning Tunneling Microscopy](https://mlsanity.com/paper/2506.01678/) (2025)
- [Spatiotemporal Satellite Image Downscaling with Transfer Encoders and Autoregressive Generative Models](https://mlsanity.com/paper/2512.05139/) (2025)
- [Provably Contractive and High-Quality Denoisers for Convergent Restoration](https://mlsanity.com/paper/2603.26168/) (2026)
- [Inpainting U-Net for seamless pedestrian-level wind prediction across urban morphologies](https://mlsanity.com/paper/2607.02560/) (2026)
- [Monitoring Post-Disaster Urban Recovery Using High-Resolution SAR Time Series and Unsupervised Learning: Evidence from the 2023 T"urkiye-Syria Earthquake](https://mlsanity.com/paper/2607.24180/) (2026)
- [DR-net-Mamba: Selective State-Space Modeling for Long-Range ECG Time-Series Denoising](https://mlsanity.com/paper/2609.35634/) (2026)
- [Stride Independent Patching for Deep Learning](https://mlsanity.com/paper/2610.12216/) (2026)

## Abstract (arXiv)

There is a limitation in the size of an image that can be processed using computationally demanding methods such as e.g. Convolutional Neural Networks (CNNs). Some imaging modalities - notably biological and medical - can result in images up to a few gigapixels in size, meaning that they have to be divided into smaller parts, or patches, for processing. However, when performing image segmentation, this may lead to undesirable artefacts, such as edge effects in the final re-combined image. We introduce windowing methods from signal processing to effectively reduce such edge effects. With the assumption that the central part of an image patch often holds richer contextual information than its sides and corners, we reconstruct the prediction by overlapping patches that are being weighted depending on 2-dimensional windows. We compare the results of four different windows: Hann, Bartlett-Hann, Triangular and a recently proposed window by Cui et al., and show that the cosine-based Hann window achieves the best improvement as measured by the Structural Similarity Index (SSIM). The proposed windowing method can be used together with any CNN model for segmentation without any modification and significantly improves network predictions.