检索结果-内蒙古大学图书馆

EL-CNN: An enhanced lightweight classification method for colorectal cancer histopathological images

BIOMEDICAL signal processing AND CONTROL 2025年 100卷

作者： Pan, Xing-Liang Hua, Bo Tong, Ke Li, Xia Luo, Jin-Long Yang, Hua Ding, Ju-Rong Sichuan Univ Sci & Engn Sch Automat & Informat Engn Zigong 643000 Peoples R China Sichuan Univ Sci & Engn Artificial Intelligence Key Lab Sichuan Prov Zigong 643000 Peoples R China Zigong Fourth Peoples Hosp Dept Gen Surg Zigong 643000 Peoples R China

Colorectal cancer (CRC) histopathological image classification is a critical part of diagnosing CRC. In this context, the classification efficiency of deep learning methods is higher than that of physicians. However, high complexity would weaken the flexibility of these methods in practice. This study proposed an enhanced lightweight convolutional neural network (EL-CNN) architecture with spatial pyramid pooling for automated classification of multi-class colorectal tissue histopathological images. In the EL-CNN, two enhanced convolutional block attention modules (ECBAM) were developed to enhance classification stability on different datasets. We compared the performance and complexity of the proposed EL-CNN with five common CNNs, as well as the effectiveness of the ECBAM with four popular attention networks, by a series of ablation studies on two publicly available datasets, including colorectal histology dataset and NCT-CRC-HE-100K. These models were evaluated by parameter size, FLOPs, accuracy, precision, sensitivity, specificity, F-1 score, and area under the curve. Specifically, the proposed EL-CNN with 3.72 M parameter size attained an accuracy of 95.24% on the colorectal histology dataset and 99.48 % on NCT-CRC-HE-100K. The proposed network keeps less complexity and better classification scores compared to most common convolutional neural networks, and even outperforms the existing state-of-the-art approaches. The comparative analysis shows that the proposed EL-CNN can take into account both robust performance and low complexity which may have significant application prospects in practical clinics.

关键词： Colorectal cancer Histopathological images Convolutional neural network Classification Enhanced lightweight model

来源：评论

学校读者我要写书评

暂无评论

On usage of an end-to-end deep neural architecture for handwritten digit string recognition

引用

signal image AND VIDEO processing 2024年第4期18卷 3009-3020页

作者： Omidi, Zahra BabaAli, Bagher Univ Tehran Sch Math Stat & Comp Sci Tehran Iran

Handwritten digit string recognition (HDSR) has received increased interest in recent years due to its vast practical applicability in both academia and industry. Approaches developed for handwritten text recognition (HTR) can be applied to HDSR if HDSR is viewed as a restricted version of HTR. It does, however, provide different challenges than HTR. For instance, a language model, which is critical for handwritten text recognition, is ineffective and cannot be employed in HDSR in general mode. The limited amount of training data is another problem influencing HDSR based on end-to-end deep learning methods. In this paper, we present a data-efficient end-to-end neural architecture for HDSR based on the HTR workflow. The proposed architecture is a gated fully convolutional network with no recurrent connections that is trained with CTC loss functions. In addition, two augmentation techniques are used to improve the model's performance. We examined our proposed model using the evaluation metrics introduced in the ICFHR 2014 competition. On the ORAND CAR-A, ORAND CAR-B, and CVL datasets, our best recognition rates are 95.41, 95.90, and 88.06%, respectively.

关键词： Handwritten digit string recognition Handwritten text recognition End-to-end deep neural network Gated convolutional neural network Data augmentation

来源：评论

学校读者我要写书评

暂无评论

RECOGNITION-GUIDED DIFFUSION MODEL FOR SCENE TEXT image SUPER-RESOLUTION 49

RECOGNITION-GUIDED DIFFUSION MODEL FOR SCENE TEXT IMAGE SUPE...

引用

49th IEEE International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Zhou, Yuxuan Gao, Liangcai Tang, Zhi Wei, Baole Peking Univ Wangxuan Inst Comp Technol Beijing Peoples R China

ISBN: (纸本)9798350344868;9798350344851

Scene Text image Super-Resolution (STISR) aims to enhance the resolution and legibility of text within low-resolution (LR) images, consequently elevating recognition accuracy in Scene Text Recognition (STR). Previous methods predominantly employ discriminative Convolutional neural Networks (CNNs) augmented with diverse forms of text guidance to address this issue. Nevertheless, they remain deficient when confronted with severely blurred images, due to their insufficient generation capability when little structural or semantic information can be extracted from original images. Therefore, we introduce RGDiffSR, a Recognition-Guided Diffusion model for scene text image Super-Resolution, which exhibits great generative diversity and fidelity even in challenging scenarios. Moreover, we propose a Recognition-Guided Denoising Network, to guide the diffusion model generating LR-consistent results through succinct semantic guidance. Experiments on the TextZoom dataset demonstrate the superiority of RGDiffSR over prior state-of-the-art methods in both text recognition accuracy and image fidelity.

关键词： scene text image super-resolution diffusion model attention mechanism scene text recognition

来源：评论

学校读者我要写书评

暂无评论

Adam golden search optimization enabled DCNN for classification of breast cancer using histopathological image

引用

BIOMEDICAL signal processing AND CONTROL 2024年 94卷

作者： Suganthi, N. Kotagiri, Srividya Thirupurasundari, Dr Vimala, S. SRM Inst Sci & Technol Dept Comp Sci & Engn Ramapuram Campus Chennai India GMR Inst Technol Dept Comp Sci & Engn Rajam 532127 Andhra Pradesh India Bharath Inst Higher Educ & Res Chennai India Prathyusha Engn Coll Dept Elect & Commun Engn Chennai Tamil Nadu India

Breast Cancer (BC) is a killing disorder, every year it kills millions of human beings. Early diagnosis is the only way to mitigate the mortality rate. Among all kinds of screening methods, medical imaging is an essential method for screening BC. Existing medical imaging alters the tissue structure and cell morphology. To overcome these limitations, histopathology image is used because it can support the decision of pathologists about the closeness or the non-appearance of a disease, as well as it can help in infection development estimation. Hence, this research develops an efficient method for BC classification using the proposed Adam Golden Search Optimization-based Deep Convolutional neural Network (AGSO-DCNN). Initially, Gaussian filter-enabled preprocessing is utilized for mitigating the noises composed in the input images. Afterwards, k-means clustering is used to feed the input images into the segmentation phase to reduce the complexity of the image. Then, to extract features like shape features, statistical features, Local Vector Patterns (LVP), and Pyramid Histogram of Oriented Gradients (PHOG) feature extraction is performed. Thereafter, the obtained features are forwarded to the multigrade BC classification stage, where DCNN is employed for classifying the image into six categories, such as apoptosis, tubule, mitosis, non-tubule, tumour nuclei, and non-tumor nuclei. DCNN is trained by the formulated AGSO mechanism, which is obtained by incorporating the Adam Optimizer and Golden Search Optimization (GSO) algorithm. Moreover, the AGSO-based DCNN technique achieved better accuracy, TPR and TNR with the values of 97.90%, 98.00%, and 98.30%, respectively.

关键词： Breast Cancer (BC) Deep Convolutional neural Network (DCNN) Golden Search Optimization (GSO) Adam Optimizer Deep Learning (DL)

来源：评论

学校读者我要写书评

暂无评论

Hir-net: a simple and effective heterogeneous image restoration network

引用

signal image AND VIDEO processing 2024年第1期18卷 773-784页

作者： Luo, Qing Liao, Yaohua Jing, Biao Gao, Xiang Chen, Wenhua Tan, Kaiwen Yunnan Power Grid Co Ltd Yuxi Power Supply Bur Yuxi Peoples R China Yunnan Power Grid Co Ltd Elect Power Res Inst Kunming Peoples R China Yunnan Power Technol Co Ltd Kunming Peoples R China Kunming Univ Sci & Technol Fac Informat Engn & Automat Kunming Peoples R China

image restoration refers to restoring the original image as much as possible from the damaged or degraded image. In recent years, deep learning-based methods have become the mainstream method for image restoration. However, the existing deep learning-based image restoration networks have the same encoder and decoder structures. The homogeneous image recovery network has limited feature representation capability, which limits its image recovery capability. In addition, the homogeneous network does not exploit the relationship between global and local features of the image well, resulting in poor quality of the recovered images. To address the above issues, we propose a heterogeneous image restoration network (HIR-Net). HIR-Net uses Transformer as an encoder to extract the global features of the image, and a convolutional neural network -based feature enhancement block is designed as a decoder to recover the local details of the image. In addition, HIR-Net introduces a new feature fusion strategy to enhance the feature representation capability of the network. This strategy performs cross-attention operations on global features extracted by the encoder and local features extracted by the decoder to achieve cross-fusion of global and local features. Compared with state-of-the-art methods, the proposed method can achieve better performance on image denoising, image draining, and underwater image enhancement.

关键词： image restoration Heterogeneous network Cross-attention Feature fusion

来源：评论

学校读者我要写书评

暂无评论

Classification of Red Blood Cells Randomly Selected from Dataset of National Institutes of Health 31

Classification of Red Blood Cells Randomly Selected from Dat...

引用

31st IEEE Conference on signal processing and Communications Applications (SIU)

作者： Bashi, Wasem Q. Ince, Erhan A. Eastern Mediterranean Univ Elect & Elect Engn Dept Famagusta North Cyprus Turkiye

ISBN: (纸本)9798350343557

Millions of people worldwide suffer from malaria, a potentially fatal disease. Early and precise diagnosis is essential for the medical condition to be successfully treated and managed. This paper employs three computer aided methods to determine percentages of red blood cells that are either parasitic or uninfected given test set(s) randomly obtained from National Institutes of Health (NIH) dataset. The three methods employed are traditional image processing, Support Vector Machine (SVM), and Convolutional neural Networks based Deep Learning (CNN-DL). The simulations were performed using a dataset that had 27,558 images of red blood cells. The traditional image processing method achieves an accuracy of 91.97%. SVM classifier using Histogram of Oriented Gradients (HOG) features had accuracy of 88.6% and with features extracted using Local Binary Patterns (LBP) accuracy had improved to 92.5%. The two previous methods were proved to be inferior when compared with the CNN- DL classification that gave an accuracy of 95.7%.

关键词： Support Vector Machine Local Binary Patterns Histogram of Oriented Gradients Convolutional neural Networks Based Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Training Deep neural Networks with HSIC and Backpropagation

Training Deep Neural Networks with HSIC and Backpropagation

引用

2024 Asia Pacific signal and Information processing Association Annual Summit and Conference

作者： Birjais, Roshan Wang, Kevin I-Kai Abdulla, Waleed Univ Auckland Auckland New Zealand

ISBN: (纸本)9798350367331;9798350367348

Deep Learning has made significant strides in recent years, particularly in supervised learning tasks, leading to the development of numerous architectures aimed at improving various aspects of model performance. Despite the effectiveness of backpropagation (BP) and stochastic gradient descent in training deep networks, these methods are often hindered by time-intensive computations, exploding and vanishing gradients, and significant memory overhead. Alternative training strategies that reduce reliance on global BP are increasingly being explored to address these limitations. This paper proposes a simple architecture that integrates Hilbert Schmidt Independence Criterion (HSIC) layers with linear layers, where the HSIC layers are trained locally, and the linear layers are optimized using global BP. This hybrid approach mitigates the drawbacks of BP while enhancing the model's ability to learn complex features across multiple layers. Our proposed model is benchmarked against existing HSIC-only models across several datasets, including MNIST, CIFAR-10, and Fashion MNIST. Results demonstrate the superior performance of our model, in terms of accuracy achieved and memory used. Additionally, we demonstrate its robustness and effectiveness when handling noisy data. The code is available at https://***/AreeBeee/***

关键词： stochastic systems

来源：评论

学校读者我要写书评

暂无评论

GALD-SE: Guided Anisotropic Lightweight Diffusion for Efficient Speech Enhancement

引用

IEEE signal processing LETTERS 2025年 32卷 426-430页

作者： Wang, Chengzhong Gu, Jianjun Yao, Dingding Li, Junfeng Yan, Yonghong Univ Chinese Acad Sci Chinese Acad Sci Inst Acoust Key Lab Speech Acoust & Content Understanding Beijing Peoples R China

Speech enhancement is designed to enhance the intelligibility and quality of speech across diverse noise conditions. Recently, diffusion models have gained lots of attention in speech enhancement area, achieving competitive results. Current diffusion-based methods blur the distribution of the signal with isotropic Gaussian noise and recover clean speech distribution from the prior. However, these methods often suffer from a substantial computational burden. We argue that the computational inefficiency partially stems from the oversight that speech enhancement is not purely a generative task;it primarily involves noise reduction and completion of missing information, while the clean clues in the original mixture do not need to be regenerated. In this paper, we propose a method that introduces noise with anisotropic guidance during the diffusion process, allowing the neural network to preserve clean clues within noisy recordings. This approach substantially reduces computational complexity while exhibiting robustness against various forms of noise and speech distortion. Experiments demonstrate that the proposed method achieves state-of-the-art results with only approximately 4.5 million parameters, a number significantly lower than that required by other diffusion methods. This effectively narrows the model size disparity between diffusion-based and predictive speech enhancement approaches. Additionally, the proposed method performs well in very noisy scenarios, demonstrating its potential for applications in highly challenging environments.

关键词： Anisotropic Speech enhancement Noise Noise measurement Diffusion processes Estimation Anisotropic magnetoresistance Training stochastic processes Spectrogram generative diffusion models anisotropic diffusion process

来源：评论

学校读者我要写书评

暂无评论

Designing Deep neural High-Density Compression Engines for Radiology images

引用

CIRCUITS SYSTEMS AND signal processing 2023年第2期42卷 643-682页

作者： Raj, Aditya Sathish, Rakshith Sarkar, Tandra Sethuraman, Ramanathan Sheet, Debdoot Indian Inst Technol Kharagpur Kharagpur W Bengal India Apollo Gleneagles Hosp Kolkata W Bengal India Intel Technol India Pvt Ltd Bangalore Karnataka India

As a speciality, radiology produces the highest volume of medical images in clinical establishments compared to other commonly employed imaging modalities like digital pathology, ophthalmic imaging, etc. Archiving this massive quantity of images with large file sizes is a major problem since the costs associated with storing medical images continue to rise with an increase in cost of electronic storage devices. One of the possible solutions is to compress them for effective storage. The prime challenge is that each modality is distinctively characterized by dynamic range and resolution of the signal and its spatial and statistical distribution. Such variations in medical images are different from camera-acquired natural scene images. Thus, conventional natural image compression algorithms such as J2K and JPEG often fail to preserve the clinically relevant details present in medical images. We address this challenge by developing a modality-specific compressor and a modality-agnostic generic decompressor implemented using a deep neural network (DNN) and capable of preserving clinically relevant image information. Architecture of the DNN is obtained through design space exploration (DSE) with the objective to feature the least computational complexity at the highest compression and a target high-quality factor, thereby leading to a low power requirement for computation. The neural compressed bitstream is further compressed using the lossless Huffman encoding to obtain a variable bit length and high-density compression (20 x -400x). Experimental validation is performed on X-ray, CT and MRI. Through quantitative measurement and clinical validation with a radiologist in the loop, we experimentally demonstrate our approach's performance superiority over traditional methods like JPEG and J2K operating at matching compression factors.

关键词： Learning-based image compression High-density radiology image compression neural architecture search Design space exploration

来源：评论

学校读者我要写书评

暂无评论

Optical fiber vibration signal recognition based on an efficient multidimensional feature extraction network

引用

APPLIED OPTICS 2024年第8期63卷 2011-2019页

作者： Du, Yuzhou Xu, Banglian Zhang, Leihong Zhang, Yiqiang Univ Shanghai Sci & Technol Coll Commun & Art Design Shanghai 200093 Peoples R China Univ Shanghai Sci & Technol Sch Opt Elect & Comp Engn Shanghai 200093 Peoples R China

In the field of optical fiber vibration signal recognition, one-dimensional signals have few features. People often used the shallow layer of a one-dimensional convolutional neural network (1D -CNN), which results in fewer features being learned by the network, leading to a poor recognition rate. There are also many complex algorithms and data processing methods, which make the whole signal recognition process more complicated. Therefore, an optical vibration signal recognition method based on an efficient multidimensional feature extraction network was proposed. Based on ResNet-50, efficient channel attention (ECA) was used to improve image features extraction ability, and a long short-term memory (LSTM) network was used to enhance the extraction of temporal features. Three different vibration signals were collected using a phase -sensitive optical time -domain reflectometry (0OTDR) optical fiber sensing system. Vibration signals were converted into 128 x 128 grayscale images, which have more effective vibration information. The experimental results show that the three types of signals can be recognized and classified effectively by the network, and the average recognition rate is 98.67%. (c) 2024 Optica Publishing Group

关键词： Deep learning Machine learning neural networks Optical fibers Optical signals Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：