Hacking with ImJoy to just get some of the concepts....

ImJoyGaussianBlurBatchHack.imjoy.html

<docs lang="markdown">
[TODO: write documentation for this plugin.]
</docs>

<config lang="json">
{
  "name": "Batch Blur",
  "type": "window",
  "tags": [],
  "ui": "",
  "version": "0.1.0",
  "cover": "",
  "description": "[TODO: describe this plugin with one sentence.]",
  "icon": "extension",
  "inputs": null,
  "outputs": null,
  "api_version": "0.1.7",
  "env": "",
  "permissions": [],
  "requirements": ["https://unpkg.com/spectre.css/dist/spectre.min.css"],
  "dependencies": ["https://gist.githubusercontent.com/tischi/2b3e87f23694471d07bc959432115b5c/raw/ImJoyGaussianBlurHack.imjoy.html"],
  "defaults": {"w": 20, "h": 10}
}
</config>

<script lang="javascript">
class ImJoyPlugin {
  async setup() {
    const p = await api.getPlugin("Gaussian Blur")
    window.batch_process = async function(){
      const ret = await api.showFileDialog({type:"directory",title:"Please select a directory with images"})
      //api.alert("You selected: " + ret.path)
      api.showMessage("Processing...")
      const image_path = "image.png"
      const output_path = "blurred_image_tmp.png"
      const sigma = 10
      // TODO: the p.process function would have to take ret.path as an input and do the looping over all images
      const out = await p.process(image_path, output_path, sigma)
      api.alert(out.sigma) 
      api.showMessage("...done!")
    }
  }

  async run(ctx) {

  }
}

api.export(new ImJoyPlugin())
</script>

<window lang="html">
  <div>
    <button class="btn" onclick="batch_process()">Batch Blur</button>
  </div>
</window>

<style lang="css">

</style>

ImJoyGaussianBlurHack.imjoy.html

<docs lang="markdown">
[TODO: write documentation for this plugin.]
</docs>

<config lang="json">
{
  "name": "Gaussian Blur",
  "type": "native-python",
  "version": "0.1.0",
  "description": "[TODO: describe this plugin with one sentence.]",
  "tags": [],
  "ui": "",
  "cover": "",
  "inputs": null,
  "outputs": null,
  "flags": [],
  "icon": "extension",
  "api_version": "0.1.7",
  "env": "",
  "permissions": [],
  "requirements": ["scikit-image"],
  "dependencies": []
}
</config>

<script lang="python">
from imjoy import api
from skimage import data
from skimage.filters import gaussian
from skimage.io import imsave
from skimage.io import imread
import base64

class ImJoyPlugin():
    def setup(self):
        api.log('initialized')

    def run(self, ctx):
        api.alert('hello world.')
        image = data.camera()
        blurred_image = gaussian( image, 10 )
        imsave("blurred-image.png",blurred_image)
        imsave("image.png",image)
        self.compare("image.png", "blurred-image.png")

    def process(self, image_path, output_path, sigma=10):
        image = imread(image_path)
        blurred_image = gaussian(image, sigma)
        imsave(output_path, blurred_image)
        api.showMessage("Blurring done!")
        return {"out":output_path,"sigma":sigma}

    def display(self, image_path):
        with open(image_path, 'rb') as f:
            data = f.read()
            result = base64.b64encode(data).decode('ascii')
            imgurl = 'data:image/png;base64,' + result
            api.createWindow(type='imjoy/image', w=12, h=15, data={"src": imgurl})

    def compare(self, image_path_a, image_path_b):
        api.createWindow(type='imjoy/image-compare', w=12, h=15, 
            data={
                "first": self.encode( image_path_a ),
                "second": self.encode( image_path_b ) })

    def encode(self, image_path):
        with open(image_path, 'rb') as f:
            data = f.read()
            result = base64.b64encode(data).decode('ascii')
            imgurl = 'data:image/png;base64,' + result
        return imgurl

api.export(ImJoyPlugin())
</script>

done

Great, thanks!