Towards detection and classification of microscopic foraminifera using transfer learning

Johansen, Thomas Haugland; Sørensen, Steffen Aagaard

doi:10.7557/18.5144

Cited by 9 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To create the ground truth, a simple, yet effective, hand-crafted object detection pipeline [14] was run on each image, which produced initial segmentation mask candidates. The pipeline consisted of two steps of Gaussian smoothing, then grayscale thresholding followed by a connected components approach to detect individual specimens.…”

Section: Plankticsmentioning

confidence: 99%

“…A typical study consists of 100-200 samples from one or several cores, and the overall time-consumption in just identifying the specimens is vast. Recently developed deep learning models show promising results towards automating parts of the identification and extraction process [10][11][12][13][14]. Figure 1 shows an example of a prepared foraminifera sample-microscopic objects spread out on a plate and photographed through a microscope.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Instance Segmentation of Microscopic Foraminifera

et al. 2021

Self Cite

View full text Add to dashboard Cite

Foraminifera are single-celled marine organisms that construct shells that remain as fossils in the marine sediments. Classifying and counting these fossils are important in paleo-oceanographic and -climatological research. However, the identification and counting process has been performed manually since the 1800s and is laborious and time-consuming. In this work, we present a deep learning-based instance segmentation model for classifying, detecting, and segmenting microscopic foraminifera. Our model is based on the Mask R-CNN architecture, using model weight parameters that have learned on the COCO detection dataset. We use a fine-tuning approach to adapt the parameters on a novel object detection dataset of more than 7000 microscopic foraminifera and sediment grains. The model achieves a (COCO-style) average precision of 0.78 on the classification and detection task, and 0.80 on the segmentation task. When the model is evaluated without challenging sediment grain images, the average precision for both tasks increases to 0.84 and 0.86, respectively. Prediction results are analyzed both quantitatively and qualitatively and discussed. Based on our findings we propose several directions for future work and conclude that our proposed model is an important step towards automating the identification and counting of microscopic foraminifera.

show abstract

Section: Plankticsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Instance Segmentation of Microscopic Foraminifera

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…To create the ground truth, a simple, yet effective, hand-crafted object detection pipeline [10] was ran on each image, which produced initial segmentation mask candidates. The pipeline consisted of two steps of Gaussian smoothing, then grayscale thresholding followed by a connected components approach to detect individual specimens.…”

Section: Dataset Curationmentioning

confidence: 99%

“…A simple illustration of the preprocessing pipeline can be seen in Figure 2. For full details, see Johansen and Sørensen [10].…”

Section: Dataset Curationmentioning

confidence: 99%

“…A typical study consists of 100-200 samples from one or several cores, and the overall time-consumption in just identifying the specimens is vast. Recently developed deep learning models show promising results towards automating parts of the identification and extraction process [6][7][8][9][10]. Figure 1 shows an example of a prepared foraminifera sample -microscopic objects spread out on a plate and photographed through a microscope.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Instance Segmentation of Microscopic Foraminifera

Johansen¹,

Sørensen²,

Møllersen³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Foraminifera are single-celled marine organisms that construct shells that remain as fossils in the marine sediments. Classifying and counting these fossils are important in e.g. paleo-oceanographic and -climatological research. However, the identification and counting process has been performed manually since the 1800s and is laborious and time-consuming. In this work, we present a deep learning-based instance segmentation model for classifying, detecting, and segmenting microscopic foraminifera. Our model is based on the Mask R-CNN architecture, using model weight parameters that have learned on the COCO detection dataset. We use a fine-tuning approach to adapt the parameters on a novel object detection dataset of more than 7000 microscopic foraminifera and sediment grains. The model achieves a (COCO-style) average precision of 0.78±0.00 on the classification and detection task, and 0.80±0.00 on the segmentation task. When the model is evaluated without challenging sediment grain images, the average precision for both tasks increases to 0.84±0.00 and 0.86±0.00, respectively. Prediction results are analyzed both quantitatively and qualitatively and discussed. Based on our findings we propose several directions for future work, and conclude that our proposed model is an important step towards automating the identification and counting of microscopic foraminifera.

show abstract