In this paper we propose a new method for on-line object segmentation through human-robot interaction. Particularly, we define three types of human gestures for object learning by the size of target objects;holding sm...
详细信息
ISBN:
(纸本)9781424466757
In this paper we propose a new method for on-line object segmentation through human-robot interaction. Particularly, we define three types of human gestures for object learning by the size of target objects;holding small objects, pointing at medium ones and contacting two corners of large ones. The regions of interest where objects are likely to be located are interpreted from those gestures and represented as rectangles in captured images. For object segmentation, we suggest a marker-based watershed segmentation method which segregates an object within a region of interest in real-time performance. Experimental results show that the segmentation quality of our method is as good as that of the grabcut algorithm, but the computational time of ours is so much faster that it is appropriate for practical applications.
The central nervous system can develop complex and deadly neoplastic growths called brain tumors. Despite being relatively uncommon in comparison to other cancers, brain tumors pose particular challenges because of th...
详细信息
The central nervous system can develop complex and deadly neoplastic growths called brain tumors. Despite being relatively uncommon in comparison to other cancers, brain tumors pose particular challenges because of their delicate anatomical placement and interactions with critical brain regions. The data are taken from TCIA (The Cancer Image Archive) and Kaggle Datasets. Images are first pre-processed using amplified median filter techniques. The pre-processed images are then segmented using the grabcut method. Feature extraction is extracted using the Shape, ABCD rule, and GLCM are the features were retrieved. The MRI images are then classified into several classes using the Bi-directional Encoder Representations from Transformers-Bidirectional Long Short Term Memory (BERT-Bi-LSTM) model. Kaggle and TICA datasets are used to simulate the proposed approach, and the results are evaluated in terms of F1-score, recall, precision and accuracy. The proposed model shows improved brain tumour identification and classification. To evaluate the expected technique's efficacy, a thorough comparison of the current techniques with preceding methods is made. The trial results showed that an efficient hybrid bert model for brain tumor classification suggested strategy provided precision of 98.65%, F1-score of 98.25%, recall of 99.25%, and accuracy of 99.75%.
Text recognition has attracted increased attention recently as a result of the complexity of natural settings and the variety of text instances. Various text or character recognition methods are introduced to distingu...
详细信息
Text recognition has attracted increased attention recently as a result of the complexity of natural settings and the variety of text instances. Various text or character recognition methods are introduced to distinguish the text from the natural scene, but existing methods struggle with the distorted and highly curved text instances. Consequently, an effective method for occluded text or character detection from object-background images is developed using the suggested elephant herding exponential sailfish optimizer-based generative adversarial network (EHESFO-based GAN). In order to build the proposed EHESFO, elephant herding optimization and Exponential SailFish Optimizer (ESFO) are merged. ESFO was created by fusing exponentially weighted moving average and SailFish Optimizer. With GAN, features extracted from the background and foreground of an image are efficiently used for image annotation and text recognition. The best features from the background and foreground images are extracted to create the optimal solution, which increases the efficacy and efficiency of text recognition. While taking the occlusion as 0.4, the proposed EHESFO-based GAN achieved higher accuracy of 98.1090% and lower error of 1.4%.
Owing to fogged image quality and similar gray scale of PET, the current main segmentation algorithm cannot give much attention to its effects and efficiency. Therefore, this paper presents a segmentation algorithm on...
详细信息
Owing to fogged image quality and similar gray scale of PET, the current main segmentation algorithm cannot give much attention to its effects and efficiency. Therefore, this paper presents a segmentation algorithm on the basis of visual saliency model of PET images. Firstly, manual operation substituted by optimized Itti visual saliency model distinguishes PET images in a fast way. Secondly, one should preprocess the salient images acquired, and then initialize the Gaussian mixture model of foreground and background area. Finally, the PET salient images are segmented by optimized grabcut algorithm, thus obtaining results. Compared with the other two algorithms, experimental results show that the proposed algorithm has some advantages in the simple operation, the efficient algorithm and the accurate results. At the same time, it effectively improves the efficiency of PET image segmentation and ensures the segmentation results.
To study the distributions of four kinds of fibers meta-aramid,poly(phenylene-1,3,4-oxadiazole)(POD),flameretardant viscose,and flame-retardant vinylon in the flame-retardant yarn,this paper discussed the preparation ...
详细信息
To study the distributions of four kinds of fibers meta-aramid,poly(phenylene-1,3,4-oxadiazole)(POD),flameretardant viscose,and flame-retardant vinylon in the flame-retardant yarn,this paper discussed the preparation method of the cross-section of flame-retardant multi-fiber blended *** to the characteristics of each component of flame-retardant blended yarn,the blended yarn was dyed via a two-bath two-step method with Argazol Red NF-3B and Cationic Blue *** the surface of the fiber was modified by the STR/I Plasma treatment *** embedding with resin,and sectioning with KD-202 Hand-operated Slicer,the crosssection image of the blended yarn was obtained by CX41RF Microscope(×400).By using the BP neural network model and the grabcut algorithm,the fibers were separated from the background,and the distribution of the fibers for each component of the flame-retardant blended yarn is *** results show that the crosssection image of flame-retardant blended yarn can be obtained quickly and easily using this method,and the distribution of each component fiber can be identified with the computer system.
In order to improve the accuracy of fabric texture defect detection,a combination of algorithm with grabcut and convolutional neural network is ***,a segmentation algorithm based on grab-cuts is used to locate and seg...
详细信息
In order to improve the accuracy of fabric texture defect detection,a combination of algorithm with grabcut and convolutional neural network is ***,a segmentation algorithm based on grab-cuts is used to locate and segment the defects in fabric images ***,the sample images are expanded to increasing the number of the training *** then,the convolutional neural network is optimized to learn the features of the fabric defects more efficiently,which make it suitable for fabric defect recognition and *** results shows that compared with other traditional algorithms,our model gets better performance with high accuracy of fabric defect detection.
A system to detect macfouling and microfouling on the ship hull is always challenging. The proposed system uses computer vision to analyze types and amount of fouling on the hull for subsequent analysis. The graphical...
详细信息
ISBN:
(纸本)9781467397254
A system to detect macfouling and microfouling on the ship hull is always challenging. The proposed system uses computer vision to analyze types and amount of fouling on the hull for subsequent analysis. The graphical-user-interface in a mobile microprocessor can be accessed and remotely controlled using platforms such as Windows and Android phone. Experiments have been tested on 40 different fouling images. The results show a mean accuracy of the fouling detection of approximately 79.97%, median of around 84.20% and a standard deviation of about 18.49% followed by the underwater experiments that give a static error of approximately 0.35% of the test area.
暂无评论