Breast cancer mortality can be prevented by only early, accurate mammography screening and diagnosis. Although CNN-based computer-aided diagnosis (CAD) systems for breast cancer have made tremendous progress recently,...
详细信息
Breast cancer mortality can be prevented by only early, accurate mammography screening and diagnosis. Although CNN-based computer-aided diagnosis (CAD) systems for breast cancer have made tremendous progress recently, accurate identification of mammography lesions is still difficult because of poor signal-to-noise ratio (SNR) and physiological features. In this manuscript, an Adaptive Fuzzy C-Means Segmentation and Deep Learning Model for Efficient Mammogram Classification Using VGG-Net (AFCM-DCNN) is proposed. The input image is given to Grey Code Approximation Pre-processing (GCAP) algorithm to enhance the quality of image by adjusting the pixel contrast. The preprocessed image is given to Adaptive Fuzzy C-Means (AFCM) algorithm and is applied in segmenting dominant regions in an input image. But in conventional AFCM technique, the centroid values get generated randomly, which consumes more computational time. Hence to enhance the performance of traditional AFCM, centroid value is optimally chosen by means of optimization algorithm. A technique for classifying images called DCNN analyses the input image and categorizes it as either benign, malignant and normal. The method extracts the features of the image and train with VGG-16 Net classifier. The neurons at the output layer have been designed to compute Class Centric Disease Support (CCDS) towards various classes. Accordingly, the mammogram class is identified towards detecting the brain tumor. The performance of the proposed method AFCM-DCNN exhibits higher accuracy of 29.3%, 25.6% and 24.6%, higher sensitivity of 15.4%, 16.6% compared with the existing methods. Therefore, in future work, hope to enhance the performance depending on transfer learning with similar data.
The bolt connection has characteristics of strong bearing capacity, easy maintenance and replacement, but the bolt connection structure often has loose failures during operation due to the detachability of the bolts. ...
详细信息
The bolt connection has characteristics of strong bearing capacity, easy maintenance and replacement, but the bolt connection structure often has loose failures during operation due to the detachability of the bolts. Bolts looseness and connection failure will not only affect the normal use of the mechanism, shorten the service life, and even cause casualties. Online monitoring and evaluation of bolt assembly tightness have attracted numerous interest. Automatic feature extraction plays a crucial role in intelligent state monitoring of mechanical systems, which can adaptively learn features from raw data and discover new state-sensitive features. A one-dimensional deep convolutional neural network (1D-DCNN) with eight convolutional layers and five pooling layers is proposed to achieve high precision in identification of bolt looseness. Firstly, the data overlap sampling is used to obtain the sufficient data so as to satisfy the requirements of 1D-DCNN. Then the 1D-DCNN carries out the process of feature extraction, feature selection and classification, which can take the free vibration signal of the bolt connection structure as input, and then fuse the feature extraction and assembly tightness classification process together to realize the intelligent detection of bolts looseness. The validity of the proposed method is verified by the data acquired from the free vibration excitation experiment of the bolt connection rotor of aero-engine. The results show that the adaptively learned features of the 1D-DCNN can represent the complex mapping relationship between the signal and the assembly state, and achieve higher accuracy than other methods.
image interpolation is an important topic in the field of imageprocessing. It is defined as the process of transforming low-resolution images into high-resolution ones using imageprocessingmethods. Recent studies o...
详细信息
image interpolation is an important topic in the field of imageprocessing. It is defined as the process of transforming low-resolution images into high-resolution ones using imageprocessingmethods. Recent studies on interpolation have shown that researchers are focusing on successful interpolation techniques that preserve edge information. Therefore, the edge detection phase plays a vital role in interpolation studies. However, these approaches typically rely on gradient-based linear computations for edge detection. On the other hand, non-linear structures that effectively simulate the human visual system have gained attention. In this study, a non-linear method was developed to detect edge information using a pixel similarity approach. Pixel similarity-based edge detection approach offers both lower computational complexity and more successful interpolation results compared to gradient-based approaches. 1D cubic interpolation was applied to the pixels identified as edges based on pixel similarity, while bicubic interpolation was applied to the remaining pixels. The algorithm was tested on 12 commonly used images and compared with various interpolation techniques. The results were evaluated using metrics such as SSIM and PSNR, as well as visual assessment. The experimental findings clearly demonstrated that the proposed method outperformed other approaches. Additionally, the method offers significant advantages, such as not requiring any parameters and having competitive computational cost.
Convolutional neural networks (CNNs) are widely popular in the field of image denoising. A large number of CNN-based denoising methods exhibit superior denoising performance in comparison with most conventional denois...
详细信息
Convolutional neural networks (CNNs) are widely popular in the field of image denoising. A large number of CNN-based denoising methods exhibit superior denoising performance in comparison with most conventional denoising schemes. However, some of these approaches extract the noise by stacking many common convolutional layers, which makes them prone to overfitting and causes more loss of image details since the erroneous extraction of non-noise features. A new multi-scale denoising network (MSDNet) is proposed for better tackling these issues, which uses the multi-scale feature information and pixel-wise correlation to effectively remove more noise from noisy images and retain more image details. The denoising effectiveness of MSDNet is specifically attributed to its three key modules, namely multi-scale progressive fusion block (MSPFB), pixel-wise attention block (PWAB) and residual learning (RL), in which MSPFB helps MSDNet capture more useful context information and reduce important information loss caused by ignoring scale inconsistency for capturing more noise from noisy images while maintaining more image details, PWAB facilitates MSDNet to selectively focus on specific image pixels or regions for further effectively capturing noise from noisy images while better preserving image details, and RL is helpful for MSDNet to better address deeper neural network training difficulties and mitigate overfitting. Experimental results demonstrate that MSDNet exhibits superior denoising and single-image deraining performance.
Passive millimeter-wave (PMMW) imaging is extensively employed in public security industries due to its privacy-safe and non-hazardous nature. Nevertheless, the quality of PMMW images is typically poor given blur and ...
详细信息
ISBN:
(纸本)9798350344868;9798350344851
Passive millimeter-wave (PMMW) imaging is extensively employed in public security industries due to its privacy-safe and non-hazardous nature. Nevertheless, the quality of PMMW images is typically poor given blur and noise. Although the end-to-end learning-based image deconvolution methods have demonstrated promising results, they are highly sensitive to minor variations during tests, making them fail to recover small yet crucial details. On the other hand, the blur kernel estimation from textureless PMMW images is also challenging, since the mainstream methods usually adopt convolutional neural networks ( CNNs) with local receptive fields to estimate a blur kernel adapted to the entire image. To this end, we propose a dataset-free method (PMMWDeconv) for blind PMMW image deconvolution by integrating the physical generative model into the deconvolution process. The proposed method mainly contains two subnetworks, kernel generator and image generator, and we leverage data consistency and global context priors to help the network in learning from the blurry PMMW image, where the post-processing and transformer are employed to fulfill these goals. Comprehensive experiments are performed to validate the performance of PMMWDeconv, and the results demonstrate that the proposed method surpasses state-of-the-art methods in both robustness and generalization.
Accurate segmentation of the liver and liver tumors has important clinical significance for subsequent treatment processes. Nevertheless, the existing U-shaped methods generally fall short in two challenges, constrain...
详细信息
Accurate segmentation of the liver and liver tumors has important clinical significance for subsequent treatment processes. Nevertheless, the existing U-shaped methods generally fall short in two challenges, constrained capability of extracting global features and inadequate ability in multi-scale feature expression, which contribute to a partial loss of high-level semantic information. To better cope with these issues, this paper proposes TULTSNet. The main contributions can be summarized into following three aspects: To begin with, we designed LocalGlobal Feature-Aware Transformer (LGFAT) and embedded it into the skip connection path to synergically leverage the inductive bias of convolution and Transformer's advantages in remote dependency modeling. Thus, the learning of context relationship is enhanced in the whole segmentation process. In addition, we designed a multi-scale adaptive fusion mechanism (MAFF) to establish correlations between cross receptive fields through dynamic feature interactions, particularly efficiently preserving the anatomical details of boundary-blurred tissues. Furthermore, the PW module is designed based on inter-layer and cross-layer features, enhancing the semantic resoluteness of the liver structure without the need to extend the network depth or reduce the resolution. Extensive experiments have been conducted in liver and liver tumor segmentation tasks, the Dice score in LiTS datasets of these two tasks reached 0.9557 and 0.8683, respectively, which demonstrate that our TULTS-Net is superior than other state-of-the-art methods.
Aiming at the problem that existing low-illumination image enhancement methods cannot simultaneously take into account multiple degrading factors, which leads to poor processing effect when applied to real scenes, dee...
详细信息
Artificial neural Networks (ANNs) are powerful models that can learn underlying nonlinear structures within data, such as images, sounds, and sentences. However, researchers have found a significant unsolved problem w...
详细信息
ISBN:
(纸本)9798331505561;9798331505554
Artificial neural Networks (ANNs) are powerful models that can learn underlying nonlinear structures within data, such as images, sounds, and sentences. However, researchers have found a significant unsolved problem with ANNs: small perturbations in input data or within the network's parameters can cause the network to output incorrect predictions or classifications. This vulnerability becomes even more dangerous as models are loaded onto special-purpose chips and computing devices that may be vulnerable to attackers. To address this issue, we investigate the effects of activation function perturbations using foundational mathematical theory within neural networks. We compare our theoretical results with two feed-forward neural networks trained and evaluated on the MNIST dataset. Our findings suggest that even subtle perturbations in activation functions and parameters can have a significant impact on the performance of ANNs. Our methods are effective at both strengthening and destroying ANNs.
With the increasing complexity of modern football tactics, how to intelligently and accurately analyze tactical changes in real-time during matches has become an important research direction. Traditional manual tactic...
详细信息
With the increasing complexity of modern football tactics, how to intelligently and accurately analyze tactical changes in real-time during matches has become an important research direction. Traditional manual tactical analysis methods are inefficient and susceptible to subjective bias. Therefore, using computer vision and deep learning technologies for tactical image recognition and analysis in football matches has gradually become a research hotspot. Convolutional neural Networks (CNNs), as a powerful imageprocessing tool, have been widely applied in video analysis and player detection. However, multi-target motion prediction and tracking management in dynamic football match scenes still face significant challenges. Existing research mainly focuses on static image analysis or simple player tracking, but the high-frequency image updates, player interactions, and occlusion issues in football matches complicate multi-target tracking. While some deep learning-based methods for multi-target detection and tracking have made progress, challenges remain, such as handling high-density player targets and improving motion trajectory prediction accuracy. To address these shortcomings, this study proposes two core techniques based on CNNs: first, multi-target motion prediction, which accurately forecasts players' future positions based on historical motion data;second, multi-target tracking management, which uses deep learning to track and manage each player's movement trajectory in real-time. Through these two techniques, this research aims to improve the realtime and accuracy of tactical analysis in football matches, providing coaches and analysts with more scientific and efficient tactical decision-making support.
image hiding aims to hide the secret data in the cover image for secure transmission. Recently, with the development of deep learning, some deep learning-based image hiding methods were proposed. However, most of them...
详细信息
image hiding aims to hide the secret data in the cover image for secure transmission. Recently, with the development of deep learning, some deep learning-based image hiding methods were proposed. However, most of them do not achieve outstanding hiding performance yet. To address this issue, we propose a new image hiding framework called CAE-NF, which consists of compressive autoencoders (CAE) and normalizing flow (NF). Specifically, CAE's encoder respectively maps the secret image and cover image into the corresponding feature vectors. image hiding and recovery can be modelled as the forward and backward processes of NF since NF is an invertible neural network. NF maps two feature vectors to a stego-image by its forward process. On the recovery side, the stego-images are mapped to two feature vectors by NF's backward process. Finally, the secret image is recovered by CAE's decoder. The proposed framework can achieve a good trade-off between the stego-image quality and recovered secret image quality, and meanwhile, improve the hiding and recovery performances. The experimental results demonstrate that the proposed framework significantly outperforms some state-of-the-art methods in terms of invisibility, security, and recovery accuracy on various datasets.
暂无评论