Pap Smear Multifocus Microscopic Image Fusion Database

This repository contains multifocus Pap smear microscopic image sequences, fusion results, and MATLAB implementations developed for region-based multifocus image fusion (MFIF) research in medical microscopy imaging applications.

Repository Contents

The repository contains:

1. Pap Smear Multifocus Image Sequences

50 Pap smear microscopic image sequences.

Each sequence contains:

15 RGB images,

different depth-of-field (DoF) conditions,

partially focused cervical cells.

Image dimensions:

320 × 240 pixels (RGB)

2. Fusion Results

Contains:

final fused images,

qualitative fusion results,

quantitative image quality evaluation results.

3. Comparative Algorithm Implementations

Contains implementations and evaluation outputs of several state-of-the-art multifocus image fusion methods employed for comparative analysis.

Dataset Characteristics

Imaging modality: Pap smear microscopy

Application:

Multifocus image fusion

Medical image reconstruction

Microscopic image enhancement

Number of image sequences:

50

Images per sequence:

15

Image format:

RGB images

Image size:

320 × 240 pixels

Comparative Algorithms Included

The proposed methodology was compared with the following state-of-the-art multifocus image fusion approaches:

Haghighat et al. — DCT-domain multifocus image fusion

Zhao et al. — Neighbor-distance-based multifocus fusion

Zhou et al. — Multi-scale weighted gradient fusion

Phamila and Amutha — DCT-based multifocus fusion

Kumar — Harmonic wavelet transform multifocus fusion

PM proposed method

The original CODE implementations were kindly provided by the respective authors.

The images provided in this dataset were generated in:

Tello-Mijares, Santiago, and Jesús Bescós. "Region-based multifocus image fusion for the precise acquisition of Pap smear images." Journal of biomedical optics 23.5 (2018): 056005-056005.