Skip to content

Commit d4384ad

Browse files
EpivitaeEpivitae
authored
Merge pull request #407 from imagej/ria
NIA, FIA, MIA, WinMan plugin sites added
2 parents 15e1050 + b77824c commit d4384ad

File tree

6 files changed

+321
-19
lines changed

6 files changed

+321
-19
lines changed

‎_pages/people/Epivitae.md‎

Lines changed: 13 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -10,8 +10,6 @@ affiliation:
1010
- Chimeric Nano Sensor (CNS) Team | https://www.cns.ac.cn/
1111
forum: Epivitae
1212
github: Epivitae
13-
stackoverflow:
14-
openhub:
1513
twitter: bio_sensor
1614
researchgate: Kui-Wang-37
1715
orcid: 0000-0002-9436-3632
@@ -20,34 +18,31 @@ scholar: RFAVjC4AAAAJ
2018

2119
I am a Postdoctoral Fellow at the Institute of Neuroscience (ION), CAS, where I lead the **[Chimeric Nano Sensor (CNS) Team](https://www.cns.ac.cn)**. My research focuses on the development of genetically encoded fluorescent biosensors and their application in studying neurochemical dynamics in live animals.
2220

23-
<br>
2421
<br>
2522

26-
Below are the open-source software tools I developed to facilitate imaging analysis.
23+
## 🔬 Biosensor Tools
2724

28-
### 🔬 Biosensor Tools
25+
I develop **Biosensor Tools**, a suite of open-source ImageJ/Fiji plugins designed for ratiometric imaging, motion correction, metadata inspection, and AI-powered denoising.
2926

30-
A collection of ImageJ/Fiji plugins for ratiometric imaging, motion correction, and workspace management.
27+
### 📥 Installation
3128

32-
<br>
33-
<br>
29+
The easiest way to install and stay up-to-date is via the **Fiji Update Site**:
3430

35-
**📥 Installation via Fiji Update Site**
36-
To install or update these plugins, please use the **Biosensor Tools** update site:
3731
1. Open Fiji and navigate to `Help > Update...`
3832
2. Click `Manage update sites`.
39-
3. Click `Add update site` and fill in:
40-
- **Name:** Biosensor Tools
41-
- **URL:** `https://sites.imagej.net/Biosensor-Tools/`
33+
3. Click `Add update site` and enter:
34+
* **Name:** `Biosensor Tools`
35+
* **URL:** `https://sites.imagej.net/Biosensor-Tools/`
4236
4. Close the window and click `Apply changes`.
4337

44-
<br>
4538
<br>
4639

47-
**🛠️ Plugin List**
40+
### 🛠️ Plugin List
4841

4942
| Plugin | Description | Links |
5043
| :--- | :--- | :--- |
51-
| **RIA-J** (Ratio Imaging Analyzer) | Publication-grade analysis for ratiometric biosensors. Features background subtraction, pixel-by-pixel ratio calculation, and IMD mode. | [Wiki](https://imagej.net/plugins/ria-j) / [GitHub](https://github.com/Epivitae/RIA-J) |
52-
| **FIA** (Fluorescence Image Analyzer) | A robust tool for image registration and motion correction. Designed to stabilize live-animal time-lapse datasets before quantification. | [GitHub](https://github.com/Epivitae/FIA) |
53-
| **WinMan** (Window Manager) | A workflow utility for efficiently managing the ImageJ workspace. Allows tiling, organizing, and controlling multiple image/result windows simultaneously. | [GitHub](https://github.com/Epivitae/WinMan) |
44+
| **RIA-J**<br>(Ratio Imaging Analyzer) | **Publication-grade ratiometric analysis.**<br>Features background subtraction, pixel-by-pixel ratio calculation, and IMD display mode. | [Wiki](https://imagej.net/plugins/ria-j) / [GitHub](https://github.com/Epivitae/RIA-J) |
45+
| **FIA**<br>(Fluorescence Image Aligner) | **Motion correction & Registration.**<br>Robustly stabilizes live-animal time-lapse datasets (Global & Dense Flow) before quantification. | [GitHub](https://github.com/Epivitae/FIA-Fluorescence-Image-Aligner) |
46+
| **NIA**<br>(Neural Inference Assistant) | **Zero-setup AI denoising.**<br>Runs deep learning models (DnCNN) directly in Java via ONNX Runtime. No Python, no CUDA required. Supports 5D hyperstacks. | [GitHub](https://github.com/Epivitae/NIA-Denoise) / [DOI](https://doi.org/10.5281/zenodo.18244343) |
47+
| **MIA**<br>(Metadata Inspection Analyzer) | **Metadata explorer.**<br>Instantly extracts hidden tags (Olympus .oir, OME-XML) and exports summaries to Excel/CSV. | [GitHub](https://github.com/Epivitae/MIA-Metadata-Inspection-Analyzer) |
48+
| **WinMan**<br>(Window Manager) | **Workspace accelerator.**<br>Efficiently manages, tiles, and filters multiple image windows to keep your workspace clean. | [GitHub](https://github.com/Epivitae/WinMan) |

‎_pages/plugins/3d-viewer/fia.md‎

Lines changed: 95 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,95 @@
1+
---
2+
title: FIA
3+
categories: [Registration, Motion Correction, Analysis]
4+
icon: /media/icons/FIA.png
5+
source-url: "https://github.com/Epivitae/FIA-Fluorescence-Image-Aligner"
6+
update-site: "Biosensor Tools"
7+
release-version: v3.1.0
8+
support-status: Active
9+
team-founders: ['@Epivitae']
10+
team-maintainers: ['@Epivitae']
11+
---
12+
{%- assign github = page.github -%}
13+
{%- assign release-version = page.release-version -%}
14+
{%- assign release-date = page.release-date -%}
15+
{%- assign dev-status = page.dev-status -%}
16+
{%- unless team-maintainers -%} {%- assign team-maintainers = page.team-maintainer -%} {%- endunless -%}
17+
18+
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.18206931.svg)](https://doi.org/10.5281/zenodo.18206931)
19+
20+
**FIA (Fluorescence Image Aligner)** is a robust motion correction plugin designed to solve the full spectrum of motion artifacts in biological time-lapse microscopy.
21+
22+
Developed by **Dr. Kui Wang** at the Center for Excellence in Brain Science and Intelligence Technology (CEBSIT), Chinese Academy of Sciences, FIA bridges the gap between high-precision rigid alignment and non-rigid deformation correction. It is specialized for Calcium Imaging (G/R-CaMP), Voltage Imaging, and long-term live cell monitoring.
23+
24+
Author: [Kui Wang](people/Epivitae), CEBSIT, CAS. For questions please use the [GitHub Issues](https://github.com/Epivitae/FIA-Fluorescence-Image-Aligner/issues) or tag @Epivitae on image.sc.
25+
26+
## Key Features
27+
28+
* **Dual-Engine Core**:
29+
* **OpenCV ECC**: Best for high-precision, sub-pixel rigid alignment (Rotation/Translation).
30+
* **Dense Flow (New Standard)**: State-of-the-art local deformation correction using Farnebäck optical flow. Handles internal tissue warping and soft tissue deformation better than rigid methods.
31+
* **Scientific Integrity (Intensity Preservation)**:
32+
* FIA guarantees that your **ΔF/F** analysis remains valid.
33+
* *Navigation*: Motion vectors are calculated using noise-reduced (CLAHE), contrast-enhanced temporary frames.
34+
* *Transport*: These vectors are applied to your **original raw data** using cubic interpolation. No artificial contrast is written to your final image.
35+
* **Smart "Super Reference"**:
36+
* In **Dense Flow** mode, FIA creates a reference anchor by averaging `N` frames. This eliminates the "floating anchor" problem caused by shot noise in single-frame references.
37+
38+
## Installation & Update
39+
40+
FIA is distributed via the **Biosensor Tools** Fiji Update Site.
41+
42+
1. Open **Fiji / ImageJ**.
43+
2. Navigate to **Help › Update...**
44+
3. Click **Manage update sites**.
45+
4. Check **Biosensor Tools** from the list.
46+
5. Click **Apply and Close**, then restart Fiji.
47+
48+
{% include notice icon="info" content="If the site is missing from the list, you can add it manually:<br>**Name:** `Biosensor Tools`<br>**URL:** `https://sites.imagej.net/Biosensor-Tools/`" %}
49+
50+
## Usage Guide
51+
52+
### 1. Mode Selection
53+
54+
FIA offers three operational modes tailored to different biological samples:
55+
56+
| Mode | Type | Best Application |
57+
| :--- | :--- | :--- |
58+
| **Global** | Rigid / Affine | **General Drift Correction.** Corrects XY translation and rotation. Recommended for behaving animal tracking or whole-brain imaging. |
59+
| **Dense Flow** | Non-Rigid (Optical Flow) | **95% of Biological Samples.** Uses *Super Reference + CLAHE*. Best for noisy fluorescence, brain slices, and in vivo imaging where local deformation occurs. |
60+
| **Elastic** | Non-Rigid (Raw) | **High-SNR Data.** Uses raw optical flow without preprocessing. Best for binary masks, artificial beads, or clean data where contrast enhancement is unnecessary. |
61+
62+
### 2. Parameter Tuning
63+
64+
The **Smart UI** adapts based on your selected mode.
65+
66+
#### For Dense Flow Mode:
67+
* **Flow WinSize**: The "field of view" for local alignment.
68+
* *Small (5)*: Captures fine jitter.
69+
* *Large (15+)*: Captures global tissue waves.
70+
* **Ref Depth**: Number of frames averaged to create the "Super Reference" (Default: 5). Higher values reduce noise in the reference anchor.
71+
* **Poly N**: Smoothing factor. *5* provides sharper motion borders, while *7* offers smoother flow fields but may blur distinct motion boundaries.
72+
73+
#### For Global Mode:
74+
* **Max Iterations**: The loop limit for Rigid/Affine calculation.
75+
* **Precision**: Convergence threshold (epsilon).
76+
77+
## Algorithms
78+
79+
FIA is built upon established computer vision algorithms to ensure reliability:
80+
81+
1. **Non-Rigid Registration**: Based on Farnebäck's polynomial expansion (2003) combined with CLAHE (Contrast Limited Adaptive Histogram Equalization) for robustness against uneven illumination.
82+
2. **Rigid Alignment**: Utilizes Parametric Image Alignment using Enhanced Correlation Coefficient (ECC) Maximization (Evangelidis & Psarakis, 2008).
83+
3. **Legacy Stabilization**: Includes a fallback engine based on the classic Lucas-Kanade optical flow.
84+
85+
## Citation & DOI
86+
87+
If you use FIA in your research, please cite the permanent Zenodo record:
88+
89+
Wang, K. (2026). FIA: Fluorescence Image Aligner - Robust Motion Correction for ImageJ/Fiji (v3.1.0). Zenodo. https://doi.org/10.5281/zenodo.18206931
90+
91+
* **Repository**: [https://github.com/Epivitae/FIA-Fluorescence-Image-Aligner](https://github.com/Epivitae/FIA-Fluorescence-Image-Aligner)
92+
* **CNS Team**: [Chimeric Nano Sensor Team](https://www.cns.ac.cn/)
93+
94+
---
95+
*Developed by Kui Wang © 2026. Part of the **Biosensor Tools** suite.*

‎_pages/plugins/3d-viewer/nia.md‎

Lines changed: 70 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,70 @@
1+
---
2+
title: NIA
3+
categories: [Denoising, Machine Learning, Restoration]
4+
icon: /media/icons/NIA.png
5+
source-url: "https://github.com/Epivitae/NIA-Denoise"
6+
update-site: "Biosensor Tools"
7+
release-version: v1.0.0
8+
support-status: Active
9+
team-founders: ['@Epivitae']
10+
team-maintainers: ['@Epivitae']
11+
---
12+
{%- assign github = page.github -%}
13+
{%- assign release-version = page.release-version -%}
14+
{%- assign release-date = page.release-date -%}
15+
{%- assign dev-status = page.dev-status -%}
16+
{%- unless team-maintainers -%} {%- assign team-maintainers = page.team-maintainer -%} {%- endunless -%}
17+
18+
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.18244343.svg)](https://doi.org/10.5281/zenodo.18244343)
19+
20+
**NIA (Neural Inference Assistant)** is a native, AI-powered denoising plugin that brings deep learning to ImageJ/Fiji without the configuration headache.
21+
22+
Developed by **Dr. Kui Wang** at the Center for Excellence in Brain Science and Intelligence Technology (CEBSIT), Chinese Academy of Sciences, NIA is built on the embedded **ONNX Runtime**. It allows biologists to run advanced denoising models directly in Java, eliminating the need for external Python environments, Conda, or complex CUDA setups.
23+
24+
Author: [Kui Wang](people/Epivitae), CEBSIT, CAS. For questions please use the [GitHub Issues](https://github.com/Epivitae/NIA-Denoise/issues) or tag @Epivitae on image.sc.
25+
26+
## Key Features
27+
28+
* **Zero Configuration**: Runs natively in Fiji. No Python installation, no Conda environments, and no dedicated GPU setup required.
29+
* **Proven Architecture**: Comes with a built-in, optimized **DnCNN** model (based on *Zhang et al., 2017*) specifically tuned for fluorescence microscopy.
30+
* **Extensible**: Supports loading custom user-trained models in the standard `.onnx` format.
31+
* **5D Hyperstack Support**: Seamlessly processes complex datasets (X, Y, Channel, Z-Slice, and Time-lapse) with auto-iterating logic.
32+
* **Smart Normalization**: Auto-detects bit-depth (8/16/32-bit) and applies consistent normalization to prevent "flickering" artifacts in time-lapse videos.
33+
34+
## Installation & Update
35+
36+
NIA is distributed via the **Biosensor Tools** Fiji Update Site.
37+
38+
1. Open **Fiji / ImageJ**.
39+
2. Navigate to **Help › Update...**
40+
3. Click **Manage update sites**.
41+
4. Check **Biosensor Tools** from the list.
42+
5. Click **Apply and Close**, then restart Fiji.
43+
44+
{% include notice icon="info" content="If the site is missing from the list, you can add it manually:<br>**Name:** `Biosensor Tools`<br>**URL:** `https://sites.imagej.net/Biosensor-Tools/`" %}
45+
46+
## Usage Guide
47+
48+
1. **Open Image**: Load your noisy image or stack in Fiji.
49+
2. **Launch**: Go to `Plugins › Biosensor Tools › NIA Denoise (AI)`.
50+
3. **Select Model**:
51+
* **Built-in (DnCNN)**: Recommended for general fluorescence microscopy (confocal/widefield).
52+
* **Custom ONNX**: Select this to load your own `.onnx` model file.
53+
4. **Run**: Click **Start Denoising**. The plugin will process the stack and output a new denoised window.
54+
55+
## Citation & References
56+
57+
If you use NIA in your research, please cite the software DOI:
58+
59+
**Wang, K. (2026).** NIA Denoise: User-Friendly AI Denoising Plugin for ImageJ/Fiji (v1.0.0). *Zenodo*. https://doi.org/10.5281/zenodo.18244343
60+
61+
* **Repository**: [https://github.com/Epivitae/NIA-Denoise](https://github.com/Epivitae/NIA-Denoise)
62+
* **CNS Team**: [Chimeric Nano Sensor Team](https://www.cns.ac.cn/)
63+
64+
### Methodology Reference
65+
The built-in model relies on the DnCNN architecture. Please also credit the original methodology:
66+
67+
> **Zhang, K., Zuo, W., Chen, Y., Meng, D., & Zhang, L. (2017).** Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising. *IEEE Transactions on Image Processing*, 26(7), 3142–3155.
68+
69+
---
70+
*Developed by Kui Wang © 2026. Part of the **Biosensor Tools** suite.*

‎_pages/plugins/3d-viewer/winman.md‎

Lines changed: 78 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,78 @@
1+
---
2+
title: WinMan
3+
categories: [Utilities, Workflow, Interface]
4+
icon: /media/icons/WinMan.png
5+
source-url: "https://github.com/Epivitae/WinMan"
6+
update-site: "Biosensor Tools"
7+
release-version: v1.0.0
8+
support-status: Active
9+
team-founders: ['@Epivitae']
10+
team-maintainers: ['@Epivitae']
11+
---
12+
{%- assign github = page.github -%}
13+
{%- assign release-version = page.release-version -%}
14+
{%- assign release-date = page.release-date -%}
15+
{%- assign dev-status = page.dev-status -%}
16+
{%- unless team-maintainers -%} {%- assign team-maintainers = page.team-maintainer -%} {%- endunless -%}
17+
18+
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.18224551.svg)](https://doi.org/10.5281/zenodo.18224551)
19+
20+
**WinMan (Window Manager & Workflow Accelerator)** is a lightweight utility designed to solve the "Desktop Explosion" problem in high-throughput microscopy workflows.
21+
22+
Developed by **Dr. Kui Wang** at the Center for Excellence in Brain Science and Intelligence Technology (CEBSIT), Chinese Academy of Sciences, WinMan acts as a digital housekeeper. It provides granular control over open windows, batch processing tools, and system resource monitoring with a modern, compact interface.
23+
24+
Author: [Kui Wang](people/Epivitae), CEBSIT, CAS. For questions please use the [GitHub Issues](https://github.com/Epivitae/WinMan/issues) or tag @Epivitae on image.sc.
25+
26+
## Key Features
27+
28+
* **Smart Filtering Engine**:
29+
* **Close Match**: Instantly close all windows containing a specific keyword (e.g., remove all `C1-` channels).
30+
* **Keep Match (Inverse Logic)**: The killer feature. Type `Final` and click **Keep Match** to close everything *except* your final results.
31+
* **Safety First**: "Close ALL" includes a confirmation dialog to prevent accidental data loss.
32+
* **Batch Image Tools**:
33+
* **Auto Contrast All**: Applies intelligent histogram stretching (`Enhance Contrast, saturated=0.35`) to **ALL** open images instantly. Eliminates repetitive manual adjustments.
34+
* **Reset Zoom All**: Resets all image windows to 100% zoom level for standardized viewing.
35+
* **System Health Monitor**:
36+
* Features an integrated **Memory Bar** showing live RAM usage.
37+
* **One-Click GC**: Clicking the memory bar triggers immediate Garbage Collection to free up system resources.
38+
39+
## Installation & Update
40+
41+
WinMan is distributed via the **Biosensor Tools** Fiji Update Site.
42+
43+
1. Open **Fiji / ImageJ**.
44+
2. Navigate to **Help › Update...**
45+
3. Click **Manage update sites**.
46+
4. Check **Biosensor Tools** from the list.
47+
5. Click **Apply and Close**, then restart Fiji.
48+
49+
{% include notice icon="info" content="If the site is missing from the list, you can add it manually:<br>**Name:** `Biosensor Tools`<br>**URL:** `https://sites.imagej.net/Biosensor-Tools/`" %}
50+
51+
## Usage Guide
52+
53+
### 1. Filter Logic Examples
54+
55+
| Function | Input Example | Action | Use Case |
56+
| :--- | :--- | :--- | :--- |
57+
| **Close Match** | `.csv` | Closes any window with ".csv" in title. | Cleaning up intermediate data tables. |
58+
| **Keep Match** | `Final` | Closes everything **EXCEPT** "Final" windows. | Isolating processed results for export. |
59+
| **Close Match** | `C2-` | Closes all Channel 2 images. | Removing reference channels after alignment. |
60+
61+
### 2. Interface Controls
62+
63+
* **Window Layout**:
64+
* **Tile**: Arranges images in a grid without overlap (Optimized algorithm).
65+
* **Cascade**: Stacks images diagonally with titles visible.
66+
* **Status Footer**: Displays operation feedback (e.g., "Closed 12 windows") and real-time Memory usage.
67+
68+
## Citation & DOI
69+
70+
If you use this software in your research, please cite the permanent Zenodo record:
71+
72+
Wang, K. (2026). WinMan: A Modern Window Manager and Workflow Accelerator for ImageJ/Fiji (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.18224551
73+
74+
* **Repository**: [https://github.com/Epivitae/WinMan](https://github.com/Epivitae/WinMan)
75+
* **CNS Team**: [Chimeric Nano Sensor Team](https://www.cns.ac.cn/)
76+
77+
---
78+
*Developed by Kui Wang © 2026. Part of the **Biosensor Tools** suite.*

0 commit comments

Comments
 (0)