The proceedings contain 39 papers. The topics discussed include: performance analysis of several CNN based models for brain MRI in tumor classification;MRI-based lumbar sagittal alignment classification system;3D mapp...
ISBN:
(纸本)9798350352368
The proceedings contain 39 papers. The topics discussed include: performance analysis of several CNN based models for brain MRI in tumor classification;MRI-based lumbar sagittal alignment classification system;3D mapping and landing zone identification in complex terrains using DSM and photogrammetry;vision language models for oil palm fresh fruit bunch ripeness classification;towards no shadow: region-based shadow compensation on low-altitude urban aerial images;comparative analysis of deep learning architectures for blood cancer classification;exploration of group and shuffle module for semantic segmentation of sea ice concentration;on handcrafted machine learning features for art authentication;and acoustic signature modelling of marine vessels in various environmental and operational conditions.
In recent years, due to Uv human exposure, the number of skin cancers 'subjects' cases have been increased, therefore, the accurate detection of malign skin cancer at early stage is considered as very crucial ...
详细信息
ISBN:
(纸本)9798350351491;9798350351484
In recent years, due to Uv human exposure, the number of skin cancers 'subjects' cases have been increased, therefore, the accurate detection of malign skin cancer at early stage is considered as very crucial for patients' therapy and to increase the survival rates. Melanomas is considered as the most frequent and dangerous type of skin cancer. Even a huge number of deep-learning (DL) and machine Learning (ML) based-classification methods have been introduced in the literature, there have been suspected cases during the clinical diagnosis of malignant lesions. This paper investigates and explores various DL-based models for an accurate diagnosis and detection of malign and benign skin lesions. Basically, Transfer learning (TL) techniques are adapted to efficient and accurate pre-trained models, mainly EfficientNet-B0-v2 and vision Transformers viT-b16, on the image-Net datasets. Furthermore, a modified Convolutional Neural Network (CNN) model have been adopted and trained from scratch. A publicly available benchmark dataset has been used in order to evaluate the proposed models 'performances and to compare their effectiveness with state-of-the-arts exiting methods. The obtained results are respectively 79,70%, 86,52%, and 86.97% respectively for CNN, EfficientNet-B0-v2, and viT-b16 models. The experiments have revealed the effectiveness of our proposed models compared to exiting DL and ML models for classification into benign and malignant skin lesions.
In this study, a machinevision control approach for a sun tracking system (STS) is designed, implemented, and performance is evaluated. The aim is to dynamically track the sun's centroid with high flexibility und...
详细信息
In this study, a machinevision control approach for a sun tracking system (STS) is designed, implemented, and performance is evaluated. The aim is to dynamically track the sun's centroid with high flexibility under low irradiation conditions due to weather conditions such as cloud cover. The STS is designed to work independently in the absence of a manual setup of the location's spatiotemporal data. The prototype used a 180 degrees FOv high-resolution camera as the primary sensor for accurate imageprocessing and adaptive control technique to regulate electrical signals to the two servo motors (pan and tilt). The NvIDIA Jetson AI-Computing Board is used for the autonomous deployment of the tracker. It was shown in the measurement that the sun's centroid tracking accuracy of the proposed tracker for Az (gamma) and Al (alpha) is 0.23 degrees and 0.66 degrees, respectively, with the Solar Position Algorithm (SPA) while 0.59 degrees and 0.65 degrees, respectively with the commercial solar tracker, STR-22G. The results graphically and statistically show that the prototype using machinevision can measure accurately and has the same tracking performance as compared with the two established measurements. The STS application based on machinevision control approach can meet the requirements for a dynamic and flexible control system for designed Parabolic Dish Solar Concentrators.
In this paper, deformation correction, feature extraction, image filtering, particle manipulation and other steps are used to achieve the relative positioning between the objects. The visually-assisted image processin...
详细信息
High-speed industrial machine-vision (Mv) applications such as surface inspection of steel sheets necessitate synchronous operation of multiple high-resolution cameras. Synchronization of cameras in the microsecond ba...
详细信息
High-speed industrial machine-vision (Mv) applications such as surface inspection of steel sheets necessitate synchronous operation of multiple high-resolution cameras. Synchronization of cameras in the microsecond band is necessary to ensure accurate frame matching while melding images together. Existing approaches for synchronization employ dedicated electronic circuits or network-time-protocol (NTP) whose accuracies are in the millisecond band. Conversely, IEEE-1508 precision-time-protocol (PTP) synchronizes computers in highly accurate industrial measurement and control networks. Synchronization algorithms using PTP involve synchronizing computers connected to cameras. Although the computers synchronize in the microsecond band, the cameras synchronize in the millisecond band. Moreover, PTP is practically not used for synchronizing multiple devices due to the high bandwidth utilization of the network. This paper proposes a temporal synchronization algorithm and framework with two-way communication with timestamps and estimates mean path delays. Unicast transmission forms the basis of the synchronization framework, so that the network utilization is minimal, thereby ensuring the necessary bandwidth is available for image transmission. Experimental results show that the proposed approach outperforms the existing methodologies with synchronization accuracies in the microsecond band.
The 2D FIR filters are simple to design which are more frequently used many more real world applications. The imageprocessing application like medical diagnosis, pattern recognition and robot vision. This paper intro...
详细信息
With the outstanding superposition and entanglement properties of quantum computing, quantum machine learning has attracted widespread attention in many fields, such as medical image analysis, password cracking, and p...
详细信息
With the outstanding superposition and entanglement properties of quantum computing, quantum machine learning has attracted widespread attention in many fields, such as medical image analysis, password cracking, and pattern recognition. Although classical machine learning is widely used and has shown great potential in medical image analysis, the bottlenecks of insufficient labeled data and low processing efficiency still exist. To overcome these challenges, massive studies combined quantum computing with machine learning to explore more advanced algorithms, which have achieved distin-guished improvements in parameter optimization, execution efficiency, and the reduction of error rates. Quantum machine learning provides new insights for the intersectional research of quantum technology and medical image analysis and contributes to the future development of medical image analysis. This review delivers an overview of the definition and taxonomy of quantum machine learning, as well as summarizes various quantum machine learning methods and their applications in medical image analy-sis over the past decade.(c) 2023 Elsevier B.v. All rights reserved.
PurposeThis study aims to introduce a vision-based model to generate droplets with auto-tuned parameters. The model can auto-adjust the inherent uncertainties and errors involved with the fabrication and operating par...
详细信息
PurposeThis study aims to introduce a vision-based model to generate droplets with auto-tuned parameters. The model can auto-adjust the inherent uncertainties and errors involved with the fabrication and operating parameters in microfluidic platform, attaining precise size and frequency of droplet ***/methodology/approachThe photolithography method is utilized to prepare the microfluidic devices used in this study, and various experiments are conducted at various flow-rate and viscosity ratios. Data for droplet shape is collected to train the artificial intelligence (AI) *** phase of droplets demonstrated a unique spring back effect in droplet size. The fully developed droplet sizes in the microchannel were modeled using least absolute shrinkage and selection operators (LASSO) regression model, Gaussian support vector machine (SvM), long short term memory (LSTM) and deep neural network models. Mean absolute percentage error (MAPE) of 0.05 and R2 = 0.93 were obtained with a deep neural network model on untrained flow data. The shape parameters of the droplets are affected by several uncontrolled parameters. These parameters are instinctively captured in the ***/valueExperimental data set is generated for varying viscosity values and flow rates. The variation of flow rate of continuous phase is observed here instead of dispersed phase. An automated computation routine is developed to read the droplet shape parameters considering the transient growth phase of droplets. The droplet size data is used to build and compare various AI models for predicting droplet sizes. A predictive model is developed, which is ready for automated closed loop control of the droplet generation.
With the characteristics of high I/O packaging density and excellent electrothermal performance, ceramic column grid array (CCGA) packaging has been widely used in highly reliable applications such as aerospace. For C...
详细信息
ISBN:
(纸本)9798400709234
With the characteristics of high I/O packaging density and excellent electrothermal performance, ceramic column grid array (CCGA) packaging has been widely used in highly reliable applications such as aerospace. For CCGA solder column, defect detection needs to be applied before it leaves the factory. The traditional manual detection method has low detection efficiency and the detect accuracy is greatly influenced by human subjective factors. Aiming at this problem, a set of algorithm consists of digital imageprocessing method, Yolov3 network and U-Net network has been combined to realize the surface and inner defect detection for CCGA solder column. The whole algorithm has been embedded into industrial software system based on Qt environment and field experiments have been applied. The experiment results show that the whole algorithm has good real-time performance and the detection accuracy is consistent with manual detection accuracy. The algorithm proposed in this paper can meet the needs of online defect detection for CCGA solder column.
A neural network is a machine learning (ML) program or model that processes information and recognizes patterns, similar to the human brain. The neural network algorithm operates by training on data to learn and enhan...
详细信息
暂无评论