AGAR a microbial colony dataset for deep learning detection

  title={AGAR a microbial colony dataset for deep learning detection},
  author={Sylwia Majchrowska and Jaroslaw Pawlowski and Grzegorz Guła and Tomasz Bonus and Agata Hanas and Adam Loch and Agnieszka S. Pawlak and Justyna Roszkowiak and Tomasz Golan and Zuzanna Drulis-Kawa},
The Annotated Germs for Automated Recognition (AGAR) dataset is an image database of microbial colonies cultured on agar plates. It contains 18 000 photos of five different microorganisms as single or mixed cultures, taken under diverse lighting conditions with two different cameras. All the images are classified into countable, uncountable, and empty, with the countable class labeled by microbiologists with colony location and species identification (336 442 colonies in total). This study… Expand
Generation of microbial colonies dataset with deep learning style transfer
It is shown that starting with only 100 real images, the method can generate data to train a detector that achieves comparable results to the same detector but trained on a real, several dozen times bigger dataset, which proves the usefulness of the method in microbe detection and segmentation. Expand
Deep neural networks approach to microbial colony detection - a comparative analysis
Three well-known deep learning approaches for object detection on the AGAR dataset are compared, namely two-stage, one-stage and transformer based neural networks, and the results may serve as a benchmark for future experiments. Expand


Code-free deep learning for multi-modality medical image classification
This study comprehensively analyse the performance and featureset of six platforms, using four representative cross-sectional and en-face medical imaging datasets to create image classification models and demonstrated uniformly higher classification performance with the optical coherence tomography modality. Expand
Deep Residual Learning for Image Recognition
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. Expand
The Most Suitable Number of Colonies on Plates for Counting 1.
Three new sets of data were collected, involving triplicate plates of fifteen 1: 1.4 serial dilutions of 65 samples of raw milk, and statistical methods were developed to analyze bias and variance and the most suitable limits on plates for counting were suggested. Expand
Multiple Anchor Learning for Visual Object Detection
A Multiple Instance Learning (MIL) approach that selects anchors and jointly optimizes the two modules of a CNN-based object detector, referred to as Multiple Anchor Learning (MAL), which improves the baseline RetinaNet with significant margins on the commonly used MS-COCO object detection benchmark and achieves new state-of-the-art detection performance compared with recent methods. Expand
Cascade R-CNN: Delving Into High Quality Object Detection
A simple implementation of the Cascade R-CNN is shown to surpass all single-model object detectors on the challenging COCO dataset, and experiments show that it is widely applicable across detector architectures, achieving consistent gains independently of the baseline detector strength. Expand
How transferable are features in deep neural networks?
This paper quantifies the generality versus specificity of neurons in each layer of a deep convolutional neural network and reports a few surprising results, including that initializing a network with transferred features from almost any number of layers can produce a boost to generalization that lingers even after fine-tuning to the target dataset. Expand
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
This work introduces a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals and further merge RPN and Fast R-CNN into a single network by sharing their convolutionAL features. Expand
Automated abnormality detection in lower extremity radiographs using deep learning
The findings show that a single CNN model can be effectively utilized for the identification of diverse abnormalities in highly variable radiographs of multiple body parts, a result that holds potential for improving patient triage and assisting with diagnostics in resource-limited settings. Expand
An integrated iterative annotation technique for easing neural network training in medical image analysis
An intuitive interface for data annotation and the display of neural network predictions within a commonly used digital pathology whole-slide viewer is created and it is demonstrated that segmentation of human and mouse renal micro compartments is repeatedly improved when humans interact with automatically generated annotations throughout the training process. Expand
Aggregated Residual Transformations for Deep Neural Networks
On the ImageNet-1K dataset, it is empirically show that even under the restricted condition of maintaining complexity, increasing cardinality is able to improve classification accuracy and is more effective than going deeper or wider when the authors increase the capacity. Expand