Browse Datasets
Sort by # Views, desc
Dry Bean
Images of 13,611 grains of 7 different registered dry beans were taken with a high-resolution camera. A total of 16 features; 12 dimensions and 4 shape forms, were obtained from the grains.
Optical Recognition of Handwritten Digits
Two versions of this database available; see folder
Rice (Cammeo and Osmancik)
A total of 3810 rice grain's images were taken for the two species, processed and feature inferences were made. 7 morphological features were obtained for each grain of rice.
Covertype
Classification of pixels into 7 forest cover types based on attributes such as elevation, aspect, slope, hillshade, soil-type, and more.
Raisin
Images of the Kecimen and Besni raisin varieties were obtained with CVS. A total of 900 raisins were used, including 450 from both varieties, and 7 morphological features were extracted.
Statlog (Landsat Satellite)
Multi-spectral values of pixels in 3x3 neighbourhoods in a satellite image, and the classification associated with the central pixel in each neighbourhood
DeFungi
DeFungi is a dataset for direct mycological examination of microscopic fungi images. The images are from superficial fungal infections caused by yeasts, moulds, or dermatophyte fungi. The images have been manually labelled into five classes and curated with a subject matter expert assistance. The images have been cropped with automated algorithms to produce the final dataset.
Ajwa or Medjool
This dataset is a balanced dataset to classify two categories of organic Saudi dates. The dataset contains three subsets: 1) a dataset containing hand-crafted features to classify 20 date fruits into two types of organic dates (Ajwa or Medjool); 2) a dataset for images of Ajwa and Medjool (200 images of the 20 aforementioned fruits); and 3) a dataset containing tabular data with features created automatically using deep learning to classify the two organic date types (Ajwa or Medjool). This study is considered as the first work in Arabic using shallow machine learning and deep learning to create accurate models for classifying organic Saudi dates, which would enable scholars, researchers, and developers to create machine learning applications for classifying Saudi dates in various forms like websites, mobile apps, microcontrollers, tiny machine learning and internet of things applications.
Image Recognition Task Execution Times in Mobile Edge Computing
Recorded task execution times for four Edge Servers submitted by edge node; node sends images to servers for image recognition tasks. The servers perform the tasks and return the results to nodes.
COREVQA
Recently, many benchmarks and datasets have been developed to evaluate Vision-Language Models (VLMs) using visual question answering (VQA) pairs, and models have shown significant accuracy improvements. However, these benchmarks rarely test the model's ability to accurately complete visual entailment, for instance, accepting or refuting a hypothesis based on the image. To address this, we propose COREVQA (Crowd Observations and Reasoning Entailment), a benchmark of 5608 image and synthetically generated true/false statement pairs, with images derived from the CrowdHuman dataset, to provoke visual entailment reasoning on challenging crowded images. Our results show that even the top-performing VLMs achieve accuracy below 80%, with other models performing substantially worse (39.98%-69.95%). This significant performance gap reveals key limitations in VLMs’ ability to reason over certain types of image–question pairs in crowded scenes.
0 to 10 of 12