# Read an image in JuliaHub with DataSets

**URL:** https://discourse.julialang.org/t/read-an-image-in-juliahub-with-datasets/131475
**Category:** General Usage
**Tags:** juliahub
**Created:** [August 8, 2025, 1:28pm UTC](https://discourse.julialang.org/t/read-an-image-in-juliahub-with-datasets/131475 "2025-08-08T13:28:02Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![cirobr](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/cirobr/32/219994_2.png) [@cirobr](https://discourse.julialang.org/u/cirobr)
#### Post date: [August 8, 2025, 1:28pm UTC](https://discourse.julialang.org/t/read-an-image-in-juliahub-with-datasets/131475/1 "2025-08-08T13:28:02Z")

</div>

Cheers,

I have a number of images in a folder, stored in the Data area in JuliaHub. Apparently, package DataSets is used to read an image in it. For instance:

```julia-auto
using DataSets
using Images
path = "ciro-rosa-583/figs"
tree=open(BlobTree, dataset(path))
imgs = tree["padded"]["images"]
imgf = imgs["2011_003256.jpg"]
jpeg_data = read(imgf)

```

The “image” in jpeg\_data is represented as a UInt8 vector. Could not find a way to convert it back to the original (512,512) RGB image.

Alternatively, is there a way to create a symlink in the project which points out to the main dataset folder? Could not find storage location in the shell.

Thanks in advance for advising.

---

<div class="post-metadata">

### Author: ![mbauman](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mbauman/32/31082_2.png) [@mbauman](https://discourse.julialang.org/u/mbauman)
#### Post date: [August 8, 2025, 2:31pm UTC](https://discourse.julialang.org/t/read-an-image-in-juliahub-with-datasets/131475/2 "2025-08-08T14:31:34Z")

</div>

Ah, I don’t think you’re reading what you think you are there. When you have a “BlobTree” like this, you need _two_ open steps — you first open the tree itself and then you can open an individual file.

I think you want something like:

```julia-auto
jpeg_data = open(Images.load, IO, imgf)

```
