File size: 2,994 Bytes
06876b7 87d577e 06876b7 87d577e 06876b7 87d577e 06876b7 87d577e e01f0dc 06876b7 739c788 162b0ae d985e13 5cc2564 d985e13 5cc2564 d985e13 5cc2564 a933da9 3144595 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
---
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': araucaria_excelsa
'1': chinese_fan_palm
'2': chinese_hackberry
'3': chinese_tallow_tree
'4': comphor_tree
'5': formosan_sweet_gum
'6': honduras_mahogany
'7': hoop_pine
'8': indian_walnut
'9': indigenous_cinnamon_tree
'10': macaranga
'11': marabutan
'12': royal_palm
'13': silk_floss_tree
'14': white_barkfig
splits:
- name: train
num_bytes: 720396640
num_examples: 224
- name: test
num_bytes: 429531935
num_examples: 132
- name: validation
num_bytes: 406231409
num_examples: 122
download_size: 1554491709
dataset_size: 1556159984
task_categories:
- image-classification
tags:
- biology
pretty_name: NTU-Stem
size_categories:
- n<1K
license: cc-by-nc-4.0
---
# Dataset Card for "NTU-Stem"
The NTU Tree Dataset is a high-resolution few-shot learning dataset of the stem images of 15 different tree species found in the National Taiwan University (NTU) campus. The dataset was collected using personal cellphones in an effort to increase familiarity with the campus’s natural beauty.
The dataset includes images of the stem of the following 15 tree species, along with their Chinese and English names:
| 中文名稱 | English Name |
|--------------|--------------------------|
| 大王椰子 | Royal Palm |
| 土肉桂 | Indigenous Cinnamon Tree |
| 大葉桃花心木 | Honduras Mahogany |
| 小葉南洋杉 | Araucaria Excelsa |
| 石栗 | Indian Walnut |
| 朴樹 | Chinese Hackberry |
| 血桐 | Macaranga |
| 垂榕 | White Barkfig |
| 肯氏南洋杉 | Hoop Pine |
| 美人樹 | Floss-silk Tree |
| 烏桕 | Chinese Tallow Tree |
| 楓香 | Formosan Sweet Gum |
| 榕樹 | Marabutan |
| 蒲葵 | Chinese Fan Palm |
| 樟樹 | Comphor Tree |
The dataset contains a total of 240 images, with each species class containing 8 to 16 training images and the remaining 8 to 10 images serving as test data. The images were captured at a resolution of approximately 3k x 3k pixels, providing high detail for the purpose of few-shot learning.
## Acknowledgements
This dataset was collected by the following students of National Taiwan University in Graduate Institute of Networking and Multimedia (GINM) and the Department of Computer Science and Information Engineering (CSIE).
Thanks to [@liswei](https://huggingface.co/liswei), [@roger0426](https://huggingface.co/roger0426), [@CYLiao1127](https://github.com/CYLiao1127), and [@j1u2l3i4a5n](https://github.com/j1u2l3i4a5n) for collecting the dataset. |