Segmentation of Enhalus acoroides seagrass from underwater images using the Mask R-CNN method

Seagrass is a Spermatophyta plant that has many roles, including as a primary producer in the food chain in the waters. Monitoring of seagrass meadows and conditions needs to be done in order to achieve a healthy marine ecosystem. The steps taken in monitoring seagrass are by detecting and segmenting it. The purpose of the study is to implement and get information about the performance of the Mask R-CNN algorithm in detecting and segmenting the Enhalus acoroides. The dataset consists of 500 Enhalus acoroides images that had gone through a color correction and labelling process. The training process was performed with the configuration of 0.001 learning rate, batch size of 4 and some image augmentation was used to avoid overfitting. The optimum weight value was obtained after conducting the learning process with 100 epochs. A confusion matrix was used to evaluate detection performance, and linear regression was used to evaluate the segmentation produced by the model. The model evaluation results showed an accuracy value of 0.9246, a precision value of 0.9507, a recall value of 0.9712 and a correlation coefficient value of 0.8771. The value indicates that the model can detect and segment the seagrass Enhalus acoroides well and accurately.

Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot

Convolutional Neural Networks are usually fitted with manually labelled data. The labelling process is very time-consuming since large datasets are required. The use of external hardware may help in some cases, but it also introduces noise to the labelled data. In this paper, we pose a new data labelling approach by using bootstrapping to increase the accuracy of the PeTra tool. PeTra allows a mobile robot to estimate people’s location in its environment by using a LIDAR sensor and a Convolutional Neural Network. PeTra has some limitations in specific situations, such as scenarios where there are not any people. We propose to use the actual PeTra release to label the LIDAR data used to fit the Convolutional Neural Network. We have evaluated the resulting system by comparing it with the previous one—where LIDAR data were labelled with a Real Time Location System. The new release increases the MCC-score by 65.97%.

The Dynamics of Pathology Dataset Creation Using Urine Cytology as an Example

<b><i>Introduction:</i></b> Dataset creation is one of the first tasks required for training AI algorithms but is underestimated in pathology. High-quality data are essential for training algorithms and data should be labelled accurately and include sufficient morphological diversity. The dynamics and challenges of labelling a urine cytology dataset using The Paris System (TPS) criteria are presented. <b><i>Methods:</i></b> 2,454 images were labelled by pathologist consensus via video conferencing over a 14-day period. During the labelling sessions, the dynamics of the labelling process were recorded. Quality assurance images were randomly selected from images labelled in previous sessions within this study and randomly distributed throughout new labelling sessions. To assess the effect of time on the labelling process, the labelled set of images was split into 2 groups according to the median relative label time and the time taken to label images and intersession agreement were assessed. <b><i>Results:</i></b> Labelling sessions ranged from 24 m 11 s to 41 m 06 s in length, with a median of 33 m 47 s. The majority of the 2,454 images were labelled as benign urothelial cells, with atypical and malignant urothelial cells more sparsely represented. The time taken to label individual images ranged from 1 s to 42 s with a median of 2.9 s. Labelling times differed significantly among categories, with the median label time for the atypical urothelial category being 7.2 s, followed by the malignant urothelial category at 3.8 s and the benign urothelial category at 2.9 s. The overall intersession agreement for quality assurance images was substantial. The level of agreement differed among classes of urothelial cells – benign and malignant urothelial cell classes showed almost perfect agreement and the atypical urothelial cell class showed moderate agreement. Image labelling times seemed to speed up, and there was no evidence of worsening of intersession agreement with session time. <b><i>Discussion/Conclusion:</i></b> Important aspects of pathology dataset creation are presented, illustrating the significant resources required for labelling a large dataset. We present evidence that the time taken to categorise urine cytology images varies by diagnosis/class. The known challenges relating to the reproducibility of the AUC (atypical) category in TPS when compared to the NHGUC (benign) or HGUC (malignant) categories is also confirmed.

Development of a Care Labelling Process for Compression Stockings Based on Natural (Cotton) Fibers

This study is to investigate to what extent the performance of compression stockings with cotton components deteriorates after repeated washing processes. Four compression stockings having at least one cotton constituent yarn and two all-nylon stockings as reference samples were produced under controlled commercial conditions. After analysing the data obtained, a care labelling process for the compression socks with cotton components was developed such that they can preserve their compression properties over successive laundering treatments.

Intra-Camera Supervised Person Re-Identification

Existing person re-identification (re-id) methods mostly exploit a large set of cross-camera identity labelled training data. This requires a tedious data collection and annotation process, leading to poor scalability in practical re-id applications. On the other hand unsupervised re-id methods do not need identity label information, but they usually suffer from much inferior and insufficient model performance. To overcome these fundamental limitations, we propose a novel person re-identification paradigm based on an idea ofindependentper-camera identity annotation. This eliminates the most time-consuming and tedious inter-camera identity labelling process, significantly reducing the amount of human annotation efforts. Consequently, it gives rise to a more scalable and more feasible setting, which we callIntra-Camera Supervised (ICS)person re-id, for which we formulate a Multi-tAsk mulTi-labEl (MATE) deep learning method. Specifically, MATE is designed for self-discovering the cross-camera identity correspondence in a per-camera multi-task inference framework. Extensive experiments demonstrate the cost-effectiveness superiority of our method over the alternative approaches on three large person re-id datasets. For example, MATE yields 88.7% rank-1 score on Market-1501 in the proposed ICS person re-id setting, significantly outperforming unsupervised learning models and closely approaching conventional fully supervised learning competitors.

Evaluation results of combined labelling of dwellings located in different countries

In the TripleA-reno project, a new combined labelling scheme was developed for dwellings. The combined labelling includes the evaluation of the energy performance, indoor environmental quality and well-being of occupants in dwellings. In this paper, the method of the TripleA-reno combined labelling scheme, the necessary calculations and measurements and the labelling process are introduced. In the TripleA-reno project, the developed combined labelling was successfully applied to different demonstration cases. The main results and experiences of the combined labelling for four demonstration cases located in Hungary, the Netherlands, Spain and Italy are presented.