hsi-preprocessing-toolkit 1.1.2.dev7__tar.gz

hsi_preprocessing_toolkit-1.1.2.dev7/PKG-INFO

@@ -0,0 +1,56 @@
+Metadata-Version: 2.3
+Name: hsi-preprocessing-toolkit
+Version: 1.1.2.dev7
+Summary: Hsi Preprocessing Toolkit
+Requires-Dist: einops>=0.8.1
+Requires-Dist: gradio>=5.35.0
+Requires-Dist: matplotlib>=3.10.3
+Requires-Dist: rasterio>=1.4.3
+Requires-Dist: rs-fusion-datasets>=0.16.0
+Requires-Dist: scipy>=1.15.3
+Requires-Python: >=3.13
+Description-Content-Type: text/markdown
+
+# HSI Preprocessing Toolkit
+
+![](asset/screenshot.jpg)
+
+A Hyperspectral Preprocessing Toolkit that
+1. Read the raw data from the HSI camera, and convert it into `.mat` file
+2. Read the `.mat` file
+3. Preview HSI, and convert it to RGB `.png` file
+4. Crop and rotate the HSI and preview in realtime
+5. Select spectrals of interest visually and save them into a `.mat` file
+
+
+
+## Usage
+1. Download from GitHub release
+2. Double click the `start.cmd` file
+
+## Credit
+1. [uv](https://docs.astral.sh/uv/) for providing a new reliable solution for Python application distribution.  
+2. `gradio` for modern Python Data Science UI 
+3. `rasterio` for remote sensing data reading
+4. `scipy`, `numpy`, `matplotlib` and `einops`
+5. For more projects, see `pyproject.toml`
+
+
+## License
+
+```text
+Copyright (C) 2025  songyz2019
+
+This program is free software: you can redistribute it and/or modify
+it under the terms of the GNU Affero General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Affero General Public License for more details.
+
+You should have received a copy of the GNU Affero General Public License
+along with this program.  If not, see <https://www.gnu.org/licenses/>.
+```

hsi_preprocessing_toolkit-1.1.2.dev7/README.md

@@ -0,0 +1,43 @@
+# HSI Preprocessing Toolkit
+
+![](asset/screenshot.jpg)
+
+A Hyperspectral Preprocessing Toolkit that
+1. Read the raw data from the HSI camera, and convert it into `.mat` file
+2. Read the `.mat` file
+3. Preview HSI, and convert it to RGB `.png` file
+4. Crop and rotate the HSI and preview in realtime
+5. Select spectrals of interest visually and save them into a `.mat` file
+
+
+
+## Usage
+1. Download from GitHub release
+2. Double click the `start.cmd` file
+
+## Credit
+1. [uv](https://docs.astral.sh/uv/) for providing a new reliable solution for Python application distribution.  
+2. `gradio` for modern Python Data Science UI 
+3. `rasterio` for remote sensing data reading
+4. `scipy`, `numpy`, `matplotlib` and `einops`
+5. For more projects, see `pyproject.toml`
+
+
+## License
+
+```text
+Copyright (C) 2025  songyz2019
+
+This program is free software: you can redistribute it and/or modify
+it under the terms of the GNU Affero General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Affero General Public License for more details.
+
+You should have received a copy of the GNU Affero General Public License
+along with this program.  If not, see <https://www.gnu.org/licenses/>.
+```

hsi_preprocessing_toolkit-1.1.2.dev7/pyproject.toml

@@ -0,0 +1,29 @@
+[project]
+name = "hsi-preprocessing-toolkit"
+version = "1.1.2.dev7"
+description = "Hsi Preprocessing Toolkit"
+readme = "README.md"
+requires-python = ">=3.13"
+dependencies = [
+    "einops>=0.8.1",
+    "gradio>=5.35.0",
+    "matplotlib>=3.10.3",
+    "rasterio>=1.4.3",
+    "rs-fusion-datasets>=0.16.0",
+    "scipy>=1.15.3",
+]
+
+[[tool.uv.index]]
+url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple"
+default = true
+
+[build-system]
+requires = ["uv_build >= 0.8.0, <0.9.0"]
+build-backend = "uv_build"
+
+[tool.hatch.build.targets.wheel]
+packages = ["hsi-preprocessing-toolkit/__main__.py"]
+
+
+[project.scripts]
+hsi_preprocessing_toolkit = "hsi_preprocessing_toolkit.__main__:main"

hsi_preprocessing_toolkit-1.1.2.dev7/src/hsi_preprocessing_toolkit/__main__.py

@@ -0,0 +1,580 @@
+from pathlib import Path
+import shutil
+from typing import Tuple, List
+import einops
+import numpy as np
+import matplotlib.pyplot as plt
+from rasterio import open as rasterio_open
+from scipy.io import savemat, loadmat
+import gradio as gr
+import gradio.utils
+from scipy.ndimage import rotate
+from rs_fusion_datasets.util.hsi2rgb import _hsi2rgb, hsi2rgb
+from jaxtyping import Float
+from enum import Enum
+from scanner_calc import scanner_calc_tab
+
+
+plt.rcParams['font.family'] = 'SimHei'
+
+i18n = gr.I18n(**{
+    'en': {
+        "title": "hsi-preprocessing-toolkit",
+        "load": "Load",
+        "upload_instructions": "**Upload one of the following formats:**\n1. One .hdr file + one raw data file without extension\n2. One .mat file",
+        "input_format": "Input Image Shape Format",
+        "data_files": "Data Files",
+        "manual_normalize": "Manual Normalize (affects preview only)",
+        "normalize_min": "Normalize Min",
+        "normalize_max": "Normalize Max",
+        "wavelength_start": "Wavelength Range Start",
+        "wavelength_end": "Wavelength Range End",
+        "processing": "Processing",
+        "crop": "Crop",
+        "top": "Top",
+        "bottom": "Bottom",
+        "left": "Left",
+        "right": "Right",
+        "rotate": "Rotate",
+        "rotate_degree": "Rotate Degree",
+        "preview": "Preview",
+        "apply_processing": "Apply Processing Effects",
+        "mat_data_type": "MAT Data Type",
+        "mat_format": "MAT Image Shape Format",
+        "mat_key": "Key of MAT file",
+        "compress_mat": "Produce Compressed MAT File",
+        "spectral_selection": "Spectral Selection",
+        "spectral_selection_help": "Click on the image to select pixels for spectral data extraction. The selected pixels will be plotted in the spectral plot below.",
+        "spectral_plot": "Spectral Plot",
+        "style": "Style",
+        "clear": "Clear",
+        "download": "Download",
+        "output_results": "Output Results",
+        "mat_file": "MAT File",
+        "info": "Info",
+        "same_as_input": "Same",
+        "auto_detect": "Auto Detect",
+    },
+    'zh-CN':{
+        "title": "高光谱图像预处理工具箱",
+        "load": "加载",
+        "upload_instructions": "**应上传以下两种格式中的一种**\n1. 同时上传一个.hdr文件 + 一个无后缀的数据文件\n2. 一个.mat文件",
+        "input_format": "输入数据形状",
+        "data_files": "数据文件",
+        "manual_normalize": "手动归一化(仅影响预览结果)",
+        "normalize_min": "归一化最小值",
+        "normalize_max": "归一化最大值",
+        "wavelength_start": "波长范围起始",
+        "wavelength_end": "波长范围结束",
+        "processing": "处理",
+        "crop": "裁切",
+        "top": "上",
+        "bottom": "下",
+        "left": "左",
+        "right": "右",
+        "rotate": "旋转",
+        "rotate_degree": "旋转角度",
+        "preview": "预览",
+        "apply_processing": "应用处理效果",
+        "mat_data_type": "mat文件数据类型",
+        "mat_format": "mat文件格式",
+        "mat_key": "mat文件的key",
+        "compress_mat": "启用mat文件压缩",
+        "spectral_selection": "光谱选择",
+        "spectral_selection_help": "点击预览图像图像中的像素进行光谱数据提取。选中的像素将在下方的光谱图中绘制。",
+        "spectral_plot": "光谱图",
+        "style": "样式",
+        "clear": "清空",
+        "download": "下载",
+        "output_results": "输出结果",
+        "mat_file": "MAT文件",
+        "info": "信息",
+        "same_as_input": "与输入相同",
+        "auto_detect": "自动检测",
+    }
+})
+
+class AppState(Enum):
+    NOT_LOADED = 0
+    LOADED = 2
+    PREVIEWED = 3
+    COVERTED = 4
+
+# Void -> HWC
+def load_data(
+    dat_files :List[gr.utils.NamedString] | None,
+    input_format, mat_key: str | None = None
+) -> Tuple[np.ndarray | None, Path | None]:
+    if dat_files is None or len(dat_files) == 0:
+        raise gr.Error("No data file provided. 请上传数据文件. Please upload a data file.")
+
+    dat_paths = [Path(f.name) for f in dat_files]
+    mat_paths = [p for p in dat_paths if p.suffix == '.mat']
+    hdr_paths = [p for p in dat_paths if p.suffix == '.hdr']
+    raw_paths = [p for p in dat_paths if p.suffix == '']
+
+    if len(mat_paths) > 0:
+        data_path = mat_paths[0]
+        mat_dat = loadmat(mat_paths[0], squeeze_me=True, mat_dtype=True, struct_as_record=False)
+        mat_keys = [x for x in mat_dat.keys() if not x.startswith('__') and not x.endswith('__')]
+        if mat_key is None or mat_key == '':
+            mat_key = mat_keys[0]
+            if len(mat_keys) >= 2:
+                gr.Warning(f"Multiple keys found: {mat_keys}. Using: {mat_key}")
+        else:
+            if mat_key not in mat_dat:
+                raise gr.Error(f"Key '{mat_key}' not found in the MAT file. Available keys: {mat_keys}")
+        data = mat_dat[mat_key]
+    
+    elif len(hdr_paths) == 0 or len(raw_paths) == 0:
+        raise gr.Error("Both .hdr and raw data files are required. Only one is provided.")
+    elif len(hdr_paths) > 0 and len(raw_paths) > 0:
+        data_path = raw_paths[0]
+        hdr_path = hdr_paths[0]
+        if hdr_path.parent != data_path.parent:
+            shutil.copy(hdr_path, data_path.with_suffix('.hdr'))
+        print(f"Loading {data_path}")
+        with rasterio_open(data_path, 'r') as src:
+            data = src.read()
+        print(f"Loaded {data_path}")
+    else:
+        raise gr.Error("Unknown file format")
+
+    if data is None:
+        raise gr.Error("Data loading failed.")
+    elif len(data.shape) != 3:
+        raise gr.Error(f"Data shape {data.shape} is not valid. Expected 3D array.")
+
+    if input_format == 'CHW':
+        data = einops.rearrange(data, 'c h w -> h w c')
+
+    return data, data_path
+
+
+# HWC -> HWC
+def process_img(img, crop_top, crop_left, crop_bottom, crop_right, rotate_deg):
+    # Rotate
+    if rotate_deg % 360 != 0:
+        img = rotate(img, angle=rotate_deg, axes=(0, 1), reshape=True)
+
+    # Crop
+    if crop_top > 0 or crop_left > 0 or crop_bottom > 0 or crop_right > 0:
+        crop_bottom = None if crop_bottom == 0 else -crop_bottom
+        crop_right  = None if crop_right == 0  else -crop_right
+        img = img[crop_top:crop_bottom, crop_left:crop_right, :]
+
+    return img
+
+# def update_ui(state_app_state: AppState):
+#     gr.Info(f"App state changed to {state_app_state.name}")
+    
+#     preview_visible = state_app_state in [AppState.LOADED, AppState.PREVIEWED, AppState.COVERTED]
+#     convert_visible = state_app_state in [AppState.PREVIEWED, AppState.COVERTED]
+#     plot_visible = state_app_state in [AppState.COVERTED]
+    
+#     return (
+#         gr.update(visible=preview_visible),
+#         gr.update(visible=convert_visible),
+#         gr.update(visible=plot_visible)
+#     )
+
+def gr_load(
+        dat_files :List[gradio.utils.NamedString] | None,
+        input_format, input_mat_key: str | None,
+        manual_normalize: bool, normalize_min: float, normalize_max: float,
+        wavelength_from: int, wavelength_to: int,
+    ):
+    data, data_path = load_data(dat_files, input_format, input_mat_key)
+    if data is None:
+        raise gr.Error("No data file provided or data loading failed.")
+    gr.Info("Loading data...")
+    if not manual_normalize:
+        rgb = hsi2rgb(data, wavelength_range=(wavelength_from, wavelength_to), input_format='HWC', output_format='HWC', to_u8np=True)
+    else:
+        rgb = _hsi2rgb( (data-normalize_min)/(normalize_max-normalize_min), wavelength=np.linspace(wavelength_from, wavelength_to, data.shape[-1]))
+        rgb = (rgb*255.0).astype(np.uint8)
+    gr.Success(f"Data loaded")
+    logging_text = str({
+        "original_shape": str(data.shape),
+        "original_data_type": str(data.dtype),
+        "original_reflection_range": [float(data.min()), float(data.max())],
+        "original_reflection_mean": float(data.mean()),
+        "original_reflection_std": float(data.std()),
+    })
+    return rgb, data, data_path, AppState.LOADED, logging_text
+
+def gr_preview(
+        state_rgb :Float[np.ndarray, 'h w c'] | None,
+        crop_top: int, crop_left: int, crop_bottom: int, crop_right: int, 
+        rotate_deg: int
+    ):
+    if state_rgb is None:
+        raise gr.Error("No data provided.")
+    return process_img(
+        state_rgb, crop_top, crop_left, crop_bottom, crop_right, rotate_deg
+    ), AppState.PREVIEWED
+
+def gr_download_selected_spectral(data, state_select_location, data_path):
+    if not state_select_location or len(state_select_location) == 0:
+        raise gr.Error("No spectral data selected. Please select at least one pixel to download")
+    
+    gr.Info("Converting ...")
+
+    dat_path = Path(data_path.name)
+    result = []
+    result_location = []
+    for (row, col) in state_select_location:
+        result.append(data[row, col,:])
+        result_location.append([row, col]) 
+    
+    result = np.array(result)
+    result_location = np.array(result_location)
+
+    mat_path = dat_path.with_stem(dat_path.stem + '_selected_spectral').with_suffix('.mat')
+    savemat(mat_path, {'data': result, 'location': result_location}, do_compression=True, format='5')
+
+    gr.Success("Done ...")
+    
+    return str(mat_path)
+
+def gr_convert(
+        data, data_path,
+        crop_top: int, crop_left: int, crop_bottom: int, crop_right: int, 
+        rotate_deg: int,
+        mat_dtype, output_format, mat_key, compress_mat: bool,
+        logging_text: str
+    ):
+    if output_format == 'same':
+        output_format = input_format
+    
+    data = process_img(
+        data, crop_top, crop_left, crop_bottom, crop_right, rotate_deg
+    )
+    
+    # Convert to mat
+    mat_file = data_path.with_stem(data_path.stem + '_converted').with_suffix('.mat')
+    if mat_dtype == "default":
+        data_processed = data
+    else:
+        data_processed = data.astype(mat_dtype)
+
+    if output_format == 'CHW':
+        mat_dat_sav = einops.rearrange(data_processed, 'h w c -> c h w')
+    else:
+        mat_dat_sav = data_processed
+    savemat(mat_file, {mat_key: mat_dat_sav}, do_compression=compress_mat, format='5')  
+    
+    info = {
+        'original_shape': str(data.shape),
+        'original_data_type': str(data.dtype),
+        'original_reflection_range': [float(data.min()), float(data.max())],
+        'original_reflection_mean': float(data.mean()),
+        'original_reflection_std': float(data.std()),
+        'output_shape': str(mat_dat_sav.shape),
+        'output_data_type': str(mat_dat_sav.dtype),
+        'output_reflection_range': [float(mat_dat_sav.min()), float(mat_dat_sav.max())],
+        'output_reflection_mean': float(mat_dat_sav.mean()),
+        'output_reflection_std': float(mat_dat_sav.std()),
+    }
+
+    logging_text += "\n".join([f"{k}: {v}" for k, v in info.items()]) + "\n"
+
+    # Gradio will handle the visibility of mat_file_output when a file path is returned
+    return data_processed, str(mat_file), logging_text, AppState.COVERTED
+
+
+def gr_on_img_clicked(evt: gr.SelectData, state_figure :tuple, data, state_select_location, wavelength_from: int, wavelength_to :int, plot_hint: str):
+    """绘制选中像素的光谱曲线"""
+    if data is None:
+        raise gr.Error("No converted data provided")
+    if state_select_location is None:
+        state_select_location = []
+    col, row = evt.index
+    state_select_location.append((row, col)) # WHY?
+
+    wavelengths = np.linspace(wavelength_from, wavelength_to, data.shape[-1])
+    # spectral_plot = pd.DataFrame(
+    #     {
+    #         "Wavelength": wavelengths,
+    #         "Reflectance": data[row, col, :]
+    #     }
+    # ) # typecheck: ignore
+
+    # fig, ax = plt.subplots(figsize=(10, 6))
+    if state_figure is None:
+        state_figure = plt.subplots(figsize=(10, 6))
+    fig, ax = state_figure
+    # for (row, col) in state_select_location:
+    #     ax.plot(wavelengths, data[row, col, :], plot_hint ,linewidth=2)
+    ax.plot(wavelengths, data[row, col, :], plot_hint ,linewidth=2)
+    ax.set_xlabel('Wavelength')
+    ax.set_ylabel('Reflectance')
+    plt.tight_layout()
+    state_figure = (fig, ax)
+    return fig, state_figure, state_select_location
+
+
+if __name__ == "__main__":
+    theme = gr.themes.Default(primary_hue='cyan').set(
+        button_primary_background_fill='#39c5bb',
+        button_primary_background_fill_hover="#30A8A0",
+    )
+
+    with gr.Blocks(title="hsi-preprocessing-toolkit", theme=theme) as demo:
+        state_app_state = gr.State(value=AppState.NOT_LOADED)
+        state_original_data = gr.State(value=None)
+        state_processed_data = gr.State(value=None)
+        state_data_path = gr.State(value=None,)
+        state_selected_location = gr.State(value=[])
+        state_original_rgb = gr.State(value=None)
+        state_spectral_figure = gr.State(value=None)
+
+
+        with gr.Tab(i18n("title"), id="hsi_preprocessing_toolkit"):
+            with gr.Row():
+                with gr.Column():
+                    with gr.Accordion(i18n("load")) as load_panel:
+                        gr.Markdown(i18n("upload_instructions"))
+                        with gr.Row():
+                            input_format = gr.Radio(
+                                label=i18n("input_format"),
+                                choices=['HWC', 'CHW'],
+                                value='CHW'
+                            )
+                            input_mat_key = gr.Textbox(
+                                label=i18n("mat_key"),
+                                value=None,
+                                placeholder=i18n("auto_detect"),
+                                visible=True,
+                            )
+                        dat_files = gr.File(
+                            label=i18n("data_files"),
+                            file_count="multiple",
+                            type="filepath",
+                        )
+                        manual_normalize = gr.Checkbox(
+                            label=i18n("manual_normalize"),
+                            value=False,
+                        )
+                        with gr.Row():
+                            normalize_min = gr.Number(
+                                label=i18n("normalize_min"),
+                                value=0,
+                                precision=1,
+                                visible=False,
+                            )
+                            normalize_max = gr.Number(
+                                label=i18n("normalize_max"),
+                                value=2**16-1,
+                                precision=1,
+                                visible=False,
+                            )
+
+                        def toggle_normalize_fields(manual_normalize):
+                            return (
+                                gr.update(visible=manual_normalize),
+                                gr.update(visible=manual_normalize),
+                            )
+                        manual_normalize.change(
+                            fn=toggle_normalize_fields,
+                            inputs=[manual_normalize],
+                            outputs=[normalize_min, normalize_max],
+                        )
+                        with gr.Row():
+                            wavelength_from = gr.Number(
+                                label=i18n("wavelength_start"),
+                                value=400,
+                                precision=1,
+                            )
+                            wavelength_to = gr.Number(
+                                label=i18n("wavelength_end"),
+                                value=1000,
+                                precision=1,
+                            )
+                        reload_btn = gr.Button(i18n("load"), variant="primary")
+                    
+                    with gr.Accordion(i18n("processing"), visible=False) as preview_panel:
+                        with gr.Column():
+                            gr.Markdown(f"### {i18n('crop')}")
+                            with gr.Row():
+                                crop_top = gr.Slider(
+                                    label=i18n("top"),
+                                    minimum=0,
+                                    maximum=0,
+                                    step=1,
+                                )
+                                crop_bottom = gr.Slider(
+                                    label=i18n("bottom"),
+                                    minimum=0,
+                                    maximum=0,
+                                    step=1,
+                                )
+                            with gr.Row():
+                                crop_left = gr.Slider(
+                                    label=i18n("left"),
+                                    minimum=0,
+                                    maximum=0,
+                                    step=1,
+                                )
+                                crop_right = gr.Slider(
+                                    label=i18n("right"),
+                                    minimum=0,
+                                    maximum=0,
+                                    step=1,
+                                )
+
+                        with gr.Column():
+                            gr.Markdown(f"### {i18n('rotate')}")
+                            rotate_deg = gr.Slider(
+                                label=i18n("rotate_degree"),
+                                minimum=-360,
+                                maximum=360,
+                                step=1,
+                                value=0
+                            )
+                        # 这个按钮因为实时预览实际上不需要了，但是暂且隐藏备用
+                        preview_btn  = gr.Button(i18n("preview"), variant="primary", visible=False)
+                        
+
+                    with gr.Accordion(i18n("apply_processing"), visible=False) as convert_panel:
+                        mat_dtype = gr.Radio(
+                            label=i18n("mat_data_type"),
+                            choices=["same", "uint8", "uint16", "float32"],
+                            value="float32"
+                        )
+                        output_format = gr.Radio(
+                            label=i18n("mat_format"),
+                            choices=[
+                                'same', 
+                                'HWC', 
+                                'CHW'
+                            ],
+                            value='same'
+                        )
+                        mat_key = gr.Text(
+                            label=i18n("mat_key"),
+                            value='data',
+                        )
+                        compress_mat = gr.Checkbox(
+                            label=i18n("compress_mat"),
+                            value=True
+                        )
+                        convert_btn  = gr.Button(i18n("apply_processing"), variant="primary")
+                    
+                    with gr.Accordion(i18n("spectral_selection"),visible=False) as plot_panel:
+                        gr.Markdown(i18n('spectral_selection_help'))
+                        spectral_plot = gr.Plot(
+                            label=i18n("spectral_plot"),
+                            visible=True,
+                            # x="Wavelength", y="Reflectance",
+                            # height=400, width=600,
+                        )
+                        plot_hint = gr.Textbox(
+                            label=i18n("style"),
+                            value='b-',
+                        )
+                        with gr.Row():
+                            clear_plot_btn = gr.Button(
+                                i18n("clear"),
+                                variant="secondary",
+                            )
+                            download_select_spectral = gr.DownloadButton(
+                                "Download",
+                                variant="primary",
+                            )
+
+                with gr.Column():
+                    with gr.Column(variant="panel"):
+                        gr.Markdown(f"## {i18n('output_results')}")   
+                        preview_img = gr.Image(label=i18n("preview"), format="png", height="auto", width="auto", interactive=False)
+                        mat_file_output = gr.File(
+                            label=i18n("mat_file"),
+                            type="filepath",
+                            interactive=False
+                        )
+                        logging_text = gr.Textbox(
+                            label=i18n("info"),
+                        )
+                    
+
+            # dat_files.upload(
+            #     fn=gr_on_file_upload,
+            #     inputs=[dat_files, input_format, manual_normalize, normalize_min, normalize_max, wavelength_from, wavelength_to],
+            #     outputs=[state_original_rgb, state_original_data, state_data_path, state_app_state]
+            # )
+            reload_btn.click(
+                fn=gr_load,
+                inputs=[dat_files, input_format, input_mat_key, manual_normalize, normalize_min, normalize_max, wavelength_from, wavelength_to],
+                outputs=[state_original_rgb, state_original_data, state_data_path, state_app_state, logging_text]
+            )
+            convert_btn.click(
+                fn=gr_convert,
+                inputs=[
+                    state_original_data, state_data_path,
+                    crop_top, crop_left, crop_bottom, crop_right,
+                    rotate_deg,
+                    mat_dtype, output_format, mat_key, compress_mat,
+                    logging_text
+                ],
+                outputs=[state_processed_data ,mat_file_output, logging_text, state_app_state]
+            )
+            for component in [crop_top, crop_left, crop_bottom, crop_right, rotate_deg]:
+                component.change(
+                    fn=gr_preview,
+                    inputs=[state_original_rgb, crop_top, crop_left, crop_bottom, crop_right, rotate_deg],
+                    outputs=[preview_img, state_app_state]
+                )
+            
+            preview_btn.click(
+                fn=gr_preview,
+                inputs=[
+                    state_original_rgb,
+                    crop_top, crop_left, crop_bottom, crop_right,
+                    rotate_deg,
+                ],
+                outputs=[preview_img, state_app_state]
+            )
+            preview_img.select(
+                fn=gr_on_img_clicked,
+                inputs=[state_spectral_figure, state_processed_data, state_selected_location, wavelength_from, wavelength_to, plot_hint],
+                outputs=[spectral_plot, state_spectral_figure, state_selected_location]
+            )
+            clear_plot_btn.click(
+                fn=lambda: (None, None, []),
+                outputs=[spectral_plot, state_spectral_figure, state_selected_location]
+            )
+            download_select_spectral.click(
+                fn=gr_download_selected_spectral,
+                inputs=[state_processed_data, state_selected_location, state_data_path],
+                outputs=[download_select_spectral]
+            )
+            state_original_data.change(
+                fn=lambda x: (
+                    gr.update(maximum=x.shape[1] if x is not None else 0),
+                    gr.update(maximum=x.shape[1] if x is not None else 0),
+                    gr.update(maximum=x.shape[0] if x is not None else 0),
+                    gr.update(maximum=x.shape[0] if x is not None else 0),
+                ),
+                inputs=[state_original_data],
+                outputs=[crop_left, crop_right, crop_top, crop_bottom]
+            )
+            state_original_rgb.change(
+                fn=lambda x:x,
+                inputs=[state_original_rgb],
+                outputs=[preview_img]
+            )
+            state_app_state.change(
+                fn=lambda x: (
+                    gr.update(visible=True, open=(x==AppState.NOT_LOADED)),
+                    gr.update(visible=x!=AppState.NOT_LOADED, open=(x in [AppState.LOADED, AppState.PREVIEWED])),
+                    gr.update(visible=x!=AppState.NOT_LOADED, open=(x in [AppState.LOADED, AppState.PREVIEWED])),
+                    gr.update(visible=x==AppState.COVERTED,   open=(x==AppState.COVERTED))
+                ),
+                inputs=[state_app_state],
+                outputs=[load_panel, preview_panel, convert_panel, plot_panel]
+            )
+        with gr.Tab("推扫参数计算"):
+            scanner_calc_tab()
+
+    demo.launch(share=False, inbrowser=True, i18n=i18n)
+

hsi_preprocessing_toolkit-1.1.2.dev7/src/hsi_preprocessing_toolkit/scanner_calc.py

@@ -0,0 +1,38 @@
+import gradio as gr
+
+def _calculate_scanner_parameters(meta_pixel_size, n_wpixel, focal, h, delta_t):
+    """All Unit should be in SI unit."""
+    v = (meta_pixel_size * h) / (focal * delta_t)
+    resolution = (meta_pixel_size * h) / focal
+    return v, resolution
+
+def calculate_scanner_parameters(meta_pixel_size, n_wpixel, focal, h, delta_t):
+    """ Use the provide unit"""
+    meta_pixel_size /= 1e6  # Convert um to m
+    focal /= 1000  # Convert mm to m
+    delta_t /= 1000  # Convert ms to s
+    v, resolution = _calculate_scanner_parameters(meta_pixel_size, n_wpixel, focal, h, delta_t)  # Convert ms to s
+    v *= 100  # Convert m/s to cm/s
+    resolution *= 100  # Convert m to cm
+    return v, resolution
+
+def scanner_calc_tab():
+    with gr.Row():
+        with gr.Column():
+            meta_pixel_size = gr.Number(label="像元尺寸 (um)", value=7.40, precision=2)
+            n_wpixel = gr.Number(label="横向空间像素数量", value=640)
+            focal = gr.Number(label="焦距 (mm)", value=8.0, precision=1)
+            h = gr.Number(label="物距(高度) (m)", value=0.432, precision=3)
+            delta_t = gr.Number(label="帧间隔 (ms)", value=100)
+            btn = gr.Button("计算", variant="primary")
+        with gr.Column():
+            v = gr.Number(label="推扫速度 (cm/s)", interactive=False, precision=2)
+            resolution = gr.Number(label="分辨率 (cm/pixel)", interactive=False, precision=3)
+
+        btn.click(
+            calculate_scanner_parameters,
+            inputs=[meta_pixel_size, n_wpixel, focal, h, delta_t],
+            outputs=[v, resolution]
+        )
+
+__all__ = ["scanner_calc_tab"]