COVID-19 Instance Segmentation Network Based on Semantic Information Generating Proposals

The CT images of lungs of COVID-19 patients have distinct pathological features, segmenting the lesion area accurately by the method of deep learning, which is of great significance for the diagnosis and treatment of COVID-19 patients. Instance segmentation has higher sensitivity and can output the Bounding Boxes of the lesion region, however, the traditional instance segmentation method is weak in the segmentation of small lesions, and there is still room for improvement in the segmentation accuracy. We propose a instance segmentation network which is called as Semantic R-CNN. Firstly, a semantic segmentation branch is added on the basis of Mask-RCNN, and utilizing the image processing tool Skimage in Python to label the connected domain for the result of semantic segmentation, extracting the rectangular boundaries of connected domain and using them as Proposals, which will replace the Regional Proposal Network in the instance segmentation. Secondly, the Atrous Spatial Pyramid Pooling is introduced into the Feature Pyramid Network, then improving the feature fusion method in FPN. Finally, the cascade method is introduced into the detection branch of the network to optimize the Proposals. Segmentation experiments were carried out on the pathological lesion segmentation data set of CC-CCII, the average accuracy of the semantic segmentation is 40.56mAP, and compared with the Mask-RCNN, it has improved by 9.98mAP. After fusing the results of semantic segmentation and instance segmentation, the Dice coefficient is 80.7%, the sensitivity is 85.8%, and compared with the Inf-Net, it has increased by 1.6% and 8.06% respectively. The proposed network has improved the segmentation accuracy and reduced the false-negatives. © 2022 ACM.

Computer-Aided Diagnosis of COVID-19 with Joint Instance Segmentation and Classification

To address the difficulty of small lesion area detection of COVID-19 patients in their lung CT images, the author has proposed an end-to-end network which using semantic segmentation to guide instance segmentation, and extending transfer learning to the classification of COVID-19 pneumonia, Common pneumonia and Normal. Firstly, in order to extract richer multi-scale features and increase the weight of low-level features, we have introduced the Atrous Spatial Pyramid Pooling(ASPP) into the Feature Pyramid Network(FPN), and proposed Multi-scale Reverse Attention Feature Pyramid Network, then having added a semantic segmentation branch to guide instance segmentation after the output of ASPP, finally, we have extracted the object category score by detector for auxiliary classification. Segmentation experiments were carried out on the dataset of CC-CCII and COVID-19 infection segmentation dataset, the mean average precision(mAP) is 39.57%, 35.36%, Compared with the COVID-CT-Mask-Net, it has improved by 5.52%, 2.33%, we also carried out classification experiments on the dataset that is from COVIDX-CT, the sensitivity and specificity of the model for detecting COVID-19 in test data are 95.88% and 98.95% respectively. Also, the sensitivity and specificity of the model for detecting Common pneumonia in test data are 98.62% and 99.25% respectively, the sensitivity and specificity of the model for detecting Normal in test data are 99.61% and 99.11% respectively, which are the best results based on this dataset and indicators, this shows that the proposed method can quickly and effectively help the clinician identify and diagnose COVID-19 patient through their lung CT images. © 2022 IEEE.

Conclusion and future direction in data mining and machine learning

Data becomes a new currency for the world. Due to COVID-19, a significantly fewer number of flights are running, and hence the scientists cannot forecast the weather accurately. The data capturing also goes low because of this smaller number of flights. Data mining techniques play a vital role in collecting data for prediction and forecasting using different machine learning techniques. Recommender systems are available at all emerging places like agriculture, admission, matchmaking, traveling, share market, housing loan, parenting, nutrition, and consultation. Cybersecurity and forensics are also very challenging domains to fight with cybercrimes. Only data can save an entity from cyber-attacks. This chapter concludes with the future direction in data mining and machine learning techniques dealing with some related issues. © 2022 Scrivener Publishing LLC. All rights reserved.

Edge-enhanced instance segmentation by grid regions of interest.

This paper focuses on the instance segmentation task. The purpose of instance segmentation is to jointly detect, classify and segment individual instances in images, so it is used to solve a large number of industrial tasks such as novel coronavirus diagnosis and autonomous driving. However, it is not easy for instance models to achieve good results in terms of both efficiency of prediction classes and segmentation results of instance edges. We propose a single-stage instance segmentation model EEMask (edge-enhanced mask), which generates grid ROIs (regions of interest) instead of proposal boxes. EEMask divides the image uniformly according to the grid and then calculates the relevance between the grids based on the distance and grayscale values. Finally, EEMask uses the grid relevance to generate grid ROIs and grid classes. In addition, we design an edge-enhanced layer, which enhances the model's ability to perceive instance edges by increasing the number of channels with higher contrast at the instance edges. There is not any additional convolutional layer overhead, so the whole process is efficient. We evaluate EEMask on a public benchmark. On average, EEMask is 17.8% faster than BlendMask with the same training schedule. EEMask achieves a mask AP score of 39.9 on the MS COCO dataset, which outperforms Mask RCNN by 7.5% and BlendMask by 3.9%.

Comparison of the Performance of Artificial Intelligence Models Depending on the Labelled Image by Different User Levels

As reconstruction and redevelopment accelerate, the generation of construction waste increases, and construction waste treatment technology is being developed accordingly, especially using artificial intelligence (AI). The majority of AI research projects fail as a consequence of poor learning data as opposed to the structure of the AI model. If data pre-processing and labeling, i.e., the processes prior to the training step, are not carried out with development purposes in mind, the desired AI model cannot be obtained. Therefore, in this study, the performance differences of the construction waste recognition model, after data pre-processing and labeling by individuals with different degrees of expertise, were analyzed with the goal of distinguishing construction waste accurately and increasing the recycling rate. According to the experimental results, it was shown that the mean average precision (mAP) of the AI model that trained on the dataset labeled by non-professionals was superior to that labeled by professionals, being 21.75 higher in the box and 26.47 in the mask, on average. This was because it was labeled using a similar method as the Microsoft Common Objects in Context (MS COCO) datasets used for You Only Look at Coefficients (YOLACT), despite them possessing different traits for construction waste. Construction waste is differentiated by texture and color;thus, we augmented the dataset by adding noise (texture) and changing the color to consider these traits. This resulted in a meaningful accuracy being achieved in 25 epochs—two fewer than the unreinforced dataset. In order to develop an AI model that recognizes construction waste, which is an atypical object, it is necessary to develop an explainable AI model, such as a reconstruction AI network, using the model’s feature map or by creating a dataset with weights added to the texture and color of the construction waste.

Instance Segmentation in Mobile Computing Environments for Identification of Specific Characteristics in Endangered Species

The segmentation of instances is a key topic in image processing and computer vision. There are numerous applications such as medical image analysis, video surveillance, image compression among others, in which its algorithms show significant results. Due to the COVID-19 pandemic, most countries have been affected mainly by their economy. In the food production sector, including fishing and aquaculture, it was no different. In this context, this research has as main objective to contribute to the 2030 Agenda for Sustainable Development suggested by the United Nations (UN), through a fish detection and segmentation model based on the framework Detectron2, optimizing the time of professionals in identifying specific characteristics of a particular species. To achieve this objective, this research seeks to facilitate the recognition of patterns of parts of the fish from the segmentation of instances and to stimulate scientific research in the area through the morphological information collection of certain species. The results present an accuracy, based on the Intersection over Union (IoU) indicator, of 88.4%, providing an effective solution for the collection of these characteristics. © 2021 IEEE