scholarly journals Instance Segmentation of Microscopic Foraminifera

2021 ◽  
Vol 11 (14) ◽  
pp. 6543
Author(s):  
Thomas Haugland Johansen ◽  
Steffen Aagaard Sørensen ◽  
Kajsa Møllersen ◽  
Fred Godtliebsen
Keyword(s):  
Counting Process ◽  
Detection Task ◽  
Fine Tuning ◽  
Average Precision ◽  
Novel Object ◽  
Proposed Model ◽  
Model Weight ◽  
Segmentation Task ◽  
Future Work ◽  
Instance Segmentation

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.

scholarly journals Instance Segmentation of Microscopic Foraminifera

2021 ◽  
Author(s):  
Thomas Haugland Johansen ◽  
Steffen Aagaard Sørensen ◽  
Kajsa Møllersen ◽  
Fred Godtliebsen
Keyword(s):  
Counting Process ◽  
Detection Task ◽  
Fine Tuning ◽  
Average Precision ◽  
Novel Object ◽  
Proposed Model ◽  
Model Weight ◽  
Segmentation Task ◽  
Future Work ◽  
Instance Segmentation

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.

scholarly journals DETECTION AND SEGMENTATION OF ENDOSCOPIC ARTEFACTS AND DISEASES USING DEEP ARCHITECTURES

2020 ◽  
Author(s):  
Nhan T. Nguyen ◽  
Dat Q. Tran ◽  
Dung B. Nguyen
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
Detection Task ◽  
Mean Average Precision ◽  
Model Ensemble ◽  
Average Precision ◽  
Endoscopic Images ◽  
Feature Pyramid ◽  
Segmentation Task ◽  
Multiple Detectors

ABSTRACTWe describe in this paper our deep learning-based approach for the EndoCV2020 challenge, which aims to detect and segment either artefacts or diseases in endoscopic images. For the detection task, we propose to train and optimize EfficientDet—a state-of-the-art detector—with different EfficientNet backbones using Focal loss. By ensembling multiple detectors, we obtain a mean average precision (mAP) of 0.2524 on EDD2020 and 0.2202 on EAD2020. For the segmentation task, two different architectures are proposed: UNet with EfficientNet-B3 encoder and Feature Pyramid Network (FPN) with dilated ResNet-50 encoder. Each of them is trained with an auxiliary classification branch. Our model ensemble reports an sscore of 0.5972 on EAD2020 and 0.701 on EDD2020, which were among the top submitters of both challenges.

scholarly journals T-MIS: Transparency Adaptation in Medical Image Segmentation

2021 ◽  
Vol 1 (1) ◽  
pp. 11-13
Author(s):  
Ayush Somani ◽  
Divij Singh ◽  
Dilip Prasad ◽  
Alexander Horsch
Keyword(s):  
Image Features ◽  
Medical Image Segmentation ◽  
Fine Tuning ◽  
User Interactions ◽  
Training Samples ◽  
Proposed Model ◽  
Segmentation Accuracy ◽  
Segmentation Task ◽  
Unseen Objects ◽  
Medical Segmentation

We often locate ourselves in a trade-off situation between what is predicted and understanding why the predictive modeling made such a prediction. This high-risk medical segmentation task is no different where we try to interpret how well has the model learned from the image features irrespective of its accuracy. We propose image-specific fine-tuning to make a deep learning model adaptive to specific medical imaging tasks. Experimental results reveal that: a) proposed model is more robust to segment previously unseen objects (negative test dataset) than state-of-the-art CNNs; b) image-specific fine-tuning with the proposed heuristics significantly enhances segmentation accuracy; and c) our model leads to accurate results with fewer user interactions and less user time than conventional interactive segmentation methods. The model successfully classified ’no polyp’ or ’no instruments’ in the image irrespective of the absence of negative data in training samples from Kvasir-seg and Kvasir-Instrument datasets.

scholarly journals A Novel Framework Based on Mask R-CNN and Histogram Thresholding for Scalable Segmentation of New and Old Rural Buildings

2021 ◽  
Vol 13 (6) ◽  
pp. 1070
Author(s):  
Ying Li ◽  
Weipan Xu ◽  
Haohui Chen ◽  
Junhao Jiang ◽  
Xun Li
Keyword(s):  
Rural Areas ◽  
Deep Neural Networks ◽  
Training Data ◽  
The Novel ◽  
Remote Sensing Images ◽  
Average Precision ◽  
Training Samples ◽  
Rural Buildings ◽  
Segmentation Task ◽  
Instance Segmentation

Mapping new and old buildings are of great significance for understanding socio-economic development in rural areas. In recent years, deep neural networks have achieved remarkable building segmentation results in high-resolution remote sensing images. However, the scarce training data and the varying geographical environments have posed challenges for scalable building segmentation. This study proposes a novel framework based on Mask R-CNN, named Histogram Thresholding Mask Region-Based Convolutional Neural Network (HTMask R-CNN), to extract new and old rural buildings even when the label is scarce. The framework adopts the result of single-object instance segmentation from the orthodox Mask R-CNN. Further, it classifies the rural buildings into new and old ones based on a dynamic grayscale threshold inferred from the result of a two-object instance segmentation task where training data is scarce. We found that the framework can extract more buildings and achieve a much higher mean Average Precision (mAP) than the orthodox Mask R-CNN model. We tested the novel framework’s performance with increasing training data and found that it converged even when the training samples were limited. This framework’s main contribution is to allow scalable segmentation by using significantly fewer training samples than traditional machine learning practices. That makes mapping China’s new and old rural buildings viable.

FINE-TUNING OF MODELLING STRATEGY TO SIMULATE THERMO-MECHANICAL BEHAVIOUR OF DOUBLE FRICTION PENDULUM SEISMIC ISOLATORS UST ESTIMATOR

2019 ◽  
Vol 3 (Special Issue on First SACEE'19) ◽  
pp. 165-172
Author(s):  
Vincenzo Bianco ◽  
Giorgio Monti ◽  
Nicola Pio Belfiore
Keyword(s):  
Multibody Dynamics ◽  
Mechanical Behaviour ◽  
Mechanical Model ◽  
Dynamic Behaviour ◽  
Experimental Testing ◽  
Fine Tuning ◽  
Proposed Model ◽  
Modelling Techniques ◽  
Friction Pendulum

The use of friction pendulum devices has recently attracted the attention of both academic and professional engineers for the protection of structures in seismic areas. Although the effectiveness of these has been shown by the experimental testing carried out worldwide, many aspects still need to be investigated for further improvement and optimisation. A thermo-mechanical model of a double friction pendulum device (based on the most recent modelling techniques adopted in multibody dynamics) is presented in this paper. The proposed model is based on the observation that sliding may not take place as ideally as is indicated in the literature. On the contrary, the fulfilment of geometrical compatibility between the constitutive bodies (during an earthquake) suggests a very peculiar dynamic behaviour composed of a continuous alternation of sticking and slipping phases. The thermo-mechanical model of a double friction pendulum device (based on the most recent modelling techniques adopted in multibody dynamics) is presented. The process of fine-tuning of the selected modelling strategy (available to date) is also described.

122 A Mechanistic Model of Growth and Amino Acid Deposition in the Pregnant Sow: Model Development, Evaluation, and Application

2021 ◽  
Vol 99 (Supplement_1) ◽  
pp. 55-56
Author(s):  
Christian D Ramirez-Camba ◽  
Crystal L Levesque
Keyword(s):  
Weight Gain ◽  
Amino Acid ◽  
Mechanistic Model ◽  
Model Development ◽  
Allantoic Fluid ◽  
Late Gestation ◽  
Maternal Body ◽  
Proposed Model ◽  
Development Evaluation ◽  
Model Weight

Abstract A mechanistic model was developed with the objective to characterize weight gain and essential amino acid (EAA) deposition in the different tissue pools that make up the pregnant sow: placenta, allantoic fluid, amniotic fluid, fetus, uterus, mammary gland, and maternal body were considered. The data used in this modelling approach were obtained from published scientific articles reporting weights, crude protein (CP), and EAA composition in the previously mentioned tissues; studies reporting not less than 5 datapoints across gestation were considered. A total of 12 scientific articles published between 1977 and 2020 were selected for the development of the model and the model was validated using 11 separate scientific papers. The model consists of three connected sub-models: protein deposition (Pd) model, weight gain model, and EAA deposition model. Weight gain, Pd, and EAA deposition curves were developed with nonparametric statistics using splines regression. The validation of the model showed a strong agreement between observed and predicted growth (r2 = 0.92, root mean square error = 3%). The proposed model also offered descriptive insights into the weight gain and Pd during gestation. The model suggests that the definition of time-dependent Pd is more accurately described as an increase in fluid deposition during mid-gestation coinciding with a reduction in Pd. In addition, due to differences in CP composition between pregnancy-related tissues and maternal body, Pd by itself may not be the best measurement criteria for the estimation of EAA requirement in pregnant sows. The proposed model also captures the negative maternal Pd that occurs in late gestation and indicates that litter size influences maternal tissue mobilization more than parity. The model predicts that the EAA requirements in early and mid-gestation are 75, 55 and 50% lower for primiparous sows than parity 2, 3 and 4+ sows, respectively, which suggest the potential benefits of parity segregated feeding.

scholarly journals Accurate Instance-Based Segmentation for Boundary Detection in Robot Grasping Application

2021 ◽  
Vol 11 (9) ◽  
pp. 4248
Author(s):  
Hong Hai Hoang ◽  
Bao Long Tran
Keyword(s):  
Object Segmentation ◽  
State Of The Art ◽  
Rapid Development ◽  
Spatial Relationship ◽  
Learning Technologies ◽  
Average Precision ◽  
Novel Approach ◽  
3D Camera ◽  
Robot Grasping ◽  
Instance Segmentation

With the rapid development of cameras and deep learning technologies, computer vision tasks such as object detection, object segmentation and object tracking are being widely applied in many fields of life. For robot grasping tasks, object segmentation aims to classify and localize objects, which helps robots to be able to pick objects accurately. The state-of-the-art instance segmentation network framework, Mask Region-Convolution Neural Network (Mask R-CNN), does not always perform an excellent accurate segmentation at the edge or border of objects. The approach using 3D camera, however, is able to extract the entire (foreground) objects easily but can be difficult or require a large amount of computation effort to classify it. We propose a novel approach, in which we combine Mask R-CNN with 3D algorithms by adding a 3D process branch for instance segmentation. Both outcomes of two branches are contemporaneously used to classify the pixels at the edge objects by dealing with the spatial relationship between edge region and mask region. We analyze the effectiveness of the method by testing with harsh cases of object positions, for example, objects are closed, overlapped or obscured by each other to focus on edge and border segmentation. Our proposed method is about 4 to 7% higher and more stable in IoU (intersection of union). This leads to a reach of 46% of mAP (mean Average Precision), which is a higher accuracy than its counterpart. The feasibility experiment shows that our method could be a remarkable promoting for the research of the grasping robot.

Directed Fine Tuning Using Feature Clustering for Instance Segmentation of Toxoplasmosis Fundus Images

2020 ◽  
Author(s):  
Dilanga Abeyrathna ◽  
Mahadevan Subramaniam ◽  
Parvathi Chundi ◽  
Murat Hasanreisoglu ◽  
Muhammad Sohail Halim ◽  
...  
Keyword(s):  
Fine Tuning ◽  
Fundus Images ◽  
Feature Clustering ◽  
Instance Segmentation

scholarly journals Valuation of Guaranteed Maximum Price Contracts in Korea Using the Collar Option Model

2018 ◽  
Vol 2018 ◽  
pp. 1-9
Author(s):  
Sanghyo Lee ◽  
Hyeongjae Jang ◽  
Yonghan Ahn
Keyword(s):  
Profit Sharing ◽  
Construction Companies ◽  
Construction Market ◽  
Positive Effects ◽  
Valuation Model ◽  
Maximum Price ◽  
Proposed Model ◽  
Option Model ◽  
Payoff Structure ◽  
Future Work

This study assessed the levels of risk that contractors may be subject to while executing a GMP contract by applying a collar option model to the case study of an apartment project in Korea and identified implications for the application of GMP contracts in Korea. The payoff structure of the GMP contract was defined based on the collar option model and a profit sharing ratio calculated to evaluate the risks involved in GMP contracts. The results showed that an increase in the GMP and a decrease in the expected cost and cost range were accompanied by a decrease in the profit sharing ratio. The proposed valuation model for GMP contracts is expected to help clients and contractors in Korea negotiate reasonable contracts as it enables the contractor to utilize the proposed model as basic data, the client to evaluate the performance of the contractor, and both parties to agree a reasonable profit sharing ratio. Implementing GMP contracts with CMR is likely to have a number of positive effects on the Korean construction market. However, in order to maximize these effects, it is necessary to have the ability to evaluate cost uncertainty. Accordingly, it is very important to analyze the factors that influence cost volatility. In future work, the various factors that have an impact on the GMP must be studied to maximize the positive effects of the framework proposed in this paper. An analysis of the effect of each factor on the change in the GMP will help Korean construction companies who are attempting to introduce GMP contracts to perform their preconstruction services effectively.

scholarly journals Different Forecasting Horizons Based Performance Analysis of Electricity Load Forecasting Using Multilayer Perceptron Neural Network

Forecasting ◽  
2021 ◽  
Vol 3 (4) ◽  
pp. 804-838
Author(s):  
Manogaran Madhiarasan ◽  
Mohamed Louzazni
Keyword(s):  
Neural Network ◽  
Power System ◽  
Multilayer Perceptron ◽  
Load Forecasting ◽  
Consumer Information ◽  
Fine Tuning ◽  
Atmospheric Parameters ◽  
Proposed Model ◽  
Electricity Load ◽  
Electricity Load Forecasting

With an uninterrupted power supply to the consumer, it is obligatory to balance the electricity generated by the electricity load. The effective planning of economic dispatch, reserve requirements, and quality power provision for accurate consumer information concerning the electricity load is needed. The burden on the power system engineers eased electricity load forecasting is essential to ensure the enhanced power system operation and planning for reliable power provision. Fickle nature, atmospheric parameters influence makes electricity load forecasting a very complex and challenging task. This paper proposed a multilayer perceptron neural network (MLPNN) with an association of recursive fine-tuning strategy-based different forecasting horizons model for electricity load forecasting. We consider the atmospheric parameters as the inputs to the proposed model, overcoming the atmospheric effect on electricity load forecasting. Hidden layers and hidden neurons based on performance investigation performed. Analyzed performance of the proposed model with other existing models; the comparative performance investigation reveals that the proposed forecasting model performs rigorous with a minimal evaluation index (mean square error (MSE) of 1.1506 × 10-05 for Dataset 1 and MSE of 4.0142 × 10-07 for Dataset 2 concern to the single hidden layer and MSE of 2.9962 × 10-07 for Dataset 1, and MSE of 1.0425 × 10-08 for Dataset 2 concern to two hidden layers based proposed model) and compared to the considered existing models. The proposed neural network possesses a good forecasting ability because we develop based on various atmospheric parameters as the input variables, which overcomes the variance. It has a generic performance capability for electricity load forecasting. The proposed model is robust and more reliable.

Sign in / Sign up

Export Citation Format

Share Document