Image Recognition Tasks Research Articles

Stress Sense: Enhanced Stress Detection and Management Via Image Processing

This project presents an innovative approach to stress detection by utilizing Convolutional Neural Networks (CNNs) to analyze emotional cues extracted from facial images. The proposed system employs CNNs, a class of deep learning models known for their efficacy in image recognition tasks, to automatically extract features from facial images. Through a combination of convolutional, pooling, and fully connected layers, the CNN learns hierarchical representations of facial expressions associated with various emotions, including those indicative of stress. The model is trained on a diverse dataset encompassing a wide range of facial expressions, allowing it to generalize well to unseen data. Transfer learning techniques may also be employed to leverage pre-trained CNN models, further enhancing performance with limited data.

A Multiscale Attention Segment Network-Based Semantic Segmentation Model for Landslide Remote Sensing Images

Landslide disasters have garnered significant attention due to their extensive devastating impact, leading to a growing emphasis on the prompt and precise identification and detection of landslides as a prominent area of research. Previous research has primarily relied on human–computer interactions and visual interpretation from remote sensing to identify landslides. However, these methods are time-consuming, labor-intensive, subjective, and have a low level of accuracy in extracting data. An essential task in deep learning, semantic segmentation, has been crucial to automated remote sensing image recognition tasks because of its end-to-end pixel-level classification capability. In this study, to mitigate the disadvantages of existing landslide detection methods, we propose a multiscale attention segment network (MsASNet) that acquires different scales of remote sensing image features, designs an encoder–decoder structure to strengthen the landslide boundary, and combines the channel attention mechanism to strengthen the feature extraction capability. The MsASNet model exhibited an average accuracy of 95.13% on the test set from Bijie’s landslide dataset, a mean accuracy of 91.45% on the test set from Chongqing’s landslide dataset, and a mean accuracy of 90.17% on the test set from Tianshui‘s landslide dataset, signifying its ability to extract landslide information efficiently and accurately in real time. Our proposed model may be used in efforts toward the prevention and control of geological disasters.

Labelling with dynamics: A data-efficient learning paradigm for medical image segmentation

The success of deep learning on image classification and recognition tasks has led to new applications in diverse contexts, including the field of medical imaging. However, two properties of deep neural networks (DNNs) may limit their future use in medical applications. The first is that DNNs require a large amount of labeled training data, and the second is that the deep learning-based models lack interpretability. In this paper, we propose and investigate a data-efficient framework for the task of general medical image segmentation. We address the two aforementioned challenges by introducing domain knowledge in the form of a strong prior into a deep learning framework. This prior is expressed by a customized dynamical system. We performed experiments on two different datasets, namely JSRT and ISIC2016 (heart and lungs segmentation on chest X-ray images and skin lesion segmentation on dermoscopy images). We have achieved competitive results using the same amount of training data compared to the state-of-the-art methods. More importantly, we demonstrate that our framework is extremely data-efficient, and it can achieve reliable results using extremely limited training data. Furthermore, the proposed method is rotationally invariant and insensitive to initialization.

Double reuses based residual network

Deep residual network (ResNet) had shown remarkable performance in image recognition tasks, due to its shorter connections between layers close to the input and those close to the output. And densely connected convolutional network (DenseNet) had further improved recognition performance by dense feature reuses. To improve the performance of residual units in ResNet and feature reuses in DenseNet, we propose a simple and effective convolutional network architecture, named double reuses based residual network (DRRNet). DRRNet improves the residual unit of ResNet, where the feature reuse connections are added to combine all feature maps from the convolutional layers to produce the residual, and uses a residual reuse path outside units to reuse all residuals as the final feature maps for classification. Residual learning used in DRRNet can alleviate the vanishing-gradient problem. The double reuses including inner-unit feature reuses and outer-unit residual reuses effectively decrease computational cost as compared with dense connections in DenseNet, and further strengthen the forward feature propagation. DRRNet is evaluated on three object recognition benchmark datasets and an object detection dataset. In comparison with the state-of-the-art, DRRNet achieves a good balance between classification accuracy and computational cost, and optimal detection performance.

A Review of Artificial Intelligence in Breast Imaging.

With the increasing dominance of artificial intelligence (AI) techniques, the important prospects for their application have extended to various medical fields, including domains such as in vitro diagnosis, intelligent rehabilitation, medical imaging, and prognosis. Breast cancer is a common malignancy that critically affects women's physical and mental health. Early breast cancer screening-through mammography, ultrasound, or magnetic resonance imaging (MRI)-can substantially improve the prognosis for breast cancer patients. AI applications have shown excellent performance in various image recognition tasks, and their use in breast cancer screening has been explored in numerous studies. This paper introduces relevant AI techniques and their applications in the field of medical imaging of the breast (mammography and ultrasound), specifically in terms of identifying, segmenting, and classifying lesions; assessing breast cancer risk; and improving image quality. Focusing on medical imaging for breast cancer, this paper also reviews related challenges and prospects for AI.

Unveiling the potential of progressive training diffusion model for defect image generation and recognition in industrial processes

Industrial defect image generation is a crucial technique for augmenting defect image data, catering to fulfill the exigencies of training data necessary for defect recognition methods. In recent years, generative adversarial networks have seen widespread use in generating defect images. However, constrained by the scarcity of training data and the intricate variability in various defect categories, the existing methods are susceptible to issues such as training instability, a dearth of diversity, and suboptimal image quality. To overcome these challenges, this paper proposes a defect image generation framework based on a progressive training diffusion model (PTDM). Firstly, this study adopts a denoising diffusion model, supplanting traditional generative adversarial networks, to mitigate issues related to training instability and the dearth of diversity observed in generated images. Secondly, a novel progressive training strategy based on the self-designed image quality evaluator is developed to efficiently generate numerous defect images while maintaining quality. Finally, extensive experiments on a steel surface defect image dataset were conducted to validate the performance of the proposed framework in defect image generation and recognition tasks.

Houses classification using vision transformer with shifted patch tokenization

Deep learning has recently made great progress leading to revolutionizing image recognition, speech recognition, and natural language processing tasks that were previously challenging to make using traditional techniques. Image classification offers a lot of potential for architectural design, even though it is rarely used to uncover new techniques. It can be used to determine the client’s preferences and design a building that satisfies those preferences. The different architectural styles based on culture, region, and time are one of the main challenges for image classification in architecture. Hence, it can be challenging for untrained clients to recognize an architectural style, and sometimes some buildings are made up of various types that are difficult to classify as a single style. This paper investigates the potential of employing state-of-art cutting-edge image classification algorithms in houses classification. In addition, the paper proposes the uses of Shifted Patch Tokenization (SPT) and Locality Self-Attention (LSA) in order to enhance the performance of Vision transformer (ViT) when trained to classify house images with a small dataset, opposed to the regular ViT which requires huge dataset in order to converge. Experimentally, these techniques proved to have a positive impact on the performance of the ViT, which reached 96.85% accuracy when SPT and LSA are employed.

A low-latency graph computer to identify metastable particles at the Large Hadron Collider for real-time analysis of potential dark matter signatures

Image recognition is a pervasive task in many information-processing environments. We present a solution to a difficult pattern recognition problem that lies at the heart of experimental particle physics. Future experiments with very high-intensity beams will produce a spray of thousands of particles in each beam-target or beam-beam collision. Recognizing the trajectories of these particles as they traverse layers of electronic sensors is a massive image recognition task that has never been accomplished in real time. We present a real-time processing solution that is implemented in a commercial field-programmable gate array using high-level synthesis. It is an unsupervised learning algorithm that uses techniques of graph computing. A prime application is the low-latency analysis of dark-matter signatures involving metastable charged particles that manifest as disappearing tracks.

Abstract PO5-07-04: Machine learning models can help physicians reduce unnecessary ultrasound guided breast biopsies

Abstract Background Breast ultrasound is widely used as a diagnostic tool, the weakness of the method though is a high false positive rate, leading to unnecessary biopsies. In recent years, machine learning (ML) is gaining attention for its excellent performance in image-recognition tasks. The objective of this study was to train and validate ML models to predict malignancy of breast masses identified by ultrasound using clinical features of the patients, image information from the ultrasound reports and attributes extracted from the images. Methods We prospectively collected clinical and ultrasonographic attributes as well as the images from 927 lesions classified as BI-RADS 2, 3, 4a, 4b, 4c, 5 and 6 submitted to percutaneous biopsy in four institutions. Images in PNG format were loaded with OpenCV library and converted to gray scale. A total of nine attributes were extracted from the annotated area. We trained and validated five ML algorithms (histogram gradient boosting, logistic regression, a pipeline of these two classifiers, stacking ensemble of several classifiers and lightgbm), randomly sampling 80% of the dataset for training and 20% for validation using a cross fold strategy. We trained and tested the models with five different combinations of the data collected and compared the performances. The model with the lowest mean errors was selected and the threshold tuned to minimize the cases that the model classified as benign and were truly cancer (“false not cancer”). Image preprocessing, attribute extraction, and traditional machine learning was performed in Python. Results The clinical characteristics of the patients and ultrasonographic attributes of the lesions withdrawn from the exam reports are described in Table 1. The biopsy results distribution across the BI-RADS classification of the ultrasound report is shown in Table 2 and are in accordance with the literature. Among the five tested models, the logistic regression was the one with the lowest mean errors and was selected to tune the thresholds minimizing “false not cancer” cases. Our target was up to 2% of this error to stay within the BI-RADS 3 classification and thus dismiss biopsy. Table 3 describes five different combinations of data used to construct the models. All these combinations had < 0.5% of "false not cancer”, way below the 2% threshold from the BI-RADS 3 category. The mean error of the models classifying benign lesions as cancer and inducing unnecessary biopsies varied from 22.7% to 23.3% depending on the combination of data used. Conclusions Machine learning models can potentially help physicians avoid unnecessary ultrasound guided breast biopsies using unsophisticated clinical features and attributes from the images missing < 0.5% of cancer cases. Table 1. Clinical characteristics of the included patients and ultrasound features of the breast lesions. Table 2. Histology (malignant or benign) distribution across BI-RADSⓇ classification. *We had five lesions missing this data. Table 3. Mean performances of the five models and parameters used to build them. Citation Format: Isabela Buzatto, Danilo Carlotti, Nilton Onari, Ana Luiza Faim, Sarah Recife, Licerio Miguel, Ruth Bonini, Liliane Silvestre, Lucas Figueira, Daniel Guimarães Tiezzi. Machine learning models can help physicians reduce unnecessary ultrasound guided breast biopsies [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-07-04.

Reservoir Computing Based on Oxygen-Vacancy-Mediated X-ray Optical Synaptic Device for Medical CT Bone Diagnosis.

Recognition and judgment of X-ray computed tomography (CT) images play a crucial role in medical diagnosis and disease prevention. However, the storage and calculation of the X-ray imaging system applied in the traditional CT diagnosis is separate, and the pathological judgment is based on doctors' experience, which will affect the timeliness and accuracy of decision-making. In this paper, a simple-structured reservoir computing network (RC) is proposed based on Ga2O3 X-ray optical synaptic devices to recognize medical skeletal CT images with high accuracy. Through oxygen vacancy engineering, Ga2O3 X-ray optical synaptic devices with adjustable photocurrent gain and a persistent photoconductivity effect were obtained. By using the Ga2O3 X-ray optical synaptic device as a reservoir, we constructed an RC network for medical skeletal CT diagnosis and verified its image recognition capability using the MNIST data set with an accuracy of 78.08%. In the elbow skeletal CT image recognition task, the recognition rate is as high as 100%. This work constructs a simple-structured RC network for X-ray image recognition, which is of great significance in applications in medical fields.

Research on Multimodal Generative Adversarial Networks in the Framework of Deep Learning

This project aims to align facial and vocal characteristics within a closely related common space through the construction of multi-modal generative adversarial networks (GANs). The project proposes a multi-modal approach grounded in visual perception, utilizing the Graph Cut algorithm to align feature components with the image features of each corresponding local context, thereby achieving adaptability in multi-modal information. To enhance the speed and accuracy of the modeling process, a regional attention strategy is integrated. Experimental results demonstrate that the proposed algorithm enhances the accuracy of image recognition tasks.

RETFound-enhanced community-based fundus disease screening: real-world evidence and decision curve analysis.

Visual impairments and blindness are major public health concerns globally. Effective eye disease screening aided by artificial intelligence (AI) is a promising countermeasure, although it is challenged by practical constraints such as poor image quality in community screening. The recently developed ophthalmic foundation model RETFound has shown higher accuracy in retinal image recognition tasks. This study developed an RETFound-enhanced deep learning (DL) model for multiple-eye disease screening using real-world images from community screenings. Our results revealed that our DL model improved the sensitivity and specificity by over 15% compared with commercial models. Our model also shows better generalisation ability than AI models developed using traditional processes. Additionally, decision curve analysis underscores the higher net benefit of employing our model in both urban and rural settings in China. These findings indicate that the RETFound-enhanced DL model can achieve a higher net benefit in community-based screening, advocating its adoption in low- and middle-income countries to address global eye health challenges.

Forming-less flexible memristor crossbar array for neuromorphic computing applications produced using low-temperature atomic layer deposition

The rapid development of wearable artificial intelligence devices based on memristor crossbar arrays has increased the demand for flexible electronics. However, fabricating crossbar arrays on flexible substrates faces inherent challenges, notably due to the complex fabrication process at low temperatures. Moreover, ensuring the stability and reliability of memristor device remains a crucial issue. To address these issues, this study introduced N-doped TaOx (N:TaOx) as a flexible memristor crossbar array fabricated via atomic layer deposition directly on a flexible substrate at an exceptionally low temperature (150 °C). The flexible PET/ITO/N:TaOx/TiN memristor device exhibited forming-less bipolar resistive switching properties with analog memory characteristics. This was accomplished by increasing the nitrogen doping concentration and effectively reducing oxygen-related defects in TaOx. The amorphous-phase flexible N:TaOx memristor demonstrated remarkable stability, enduring 500 switching cycles and retaining its state for 24 h during bending tests involving 104 bending cycles at 2.5 mm bending radius. Furthermore, a 6 × 6 flexible memristor crossbar array was fabricated successfully, with all 36 devices exhibiting well-defined memristor behavior and minimal variation attributed to the film's uniformity achieved through ALD process. Additionally, by programming the tunable conductance of each device in the array, characters of desired shapes can be formed and read. Regarding their synaptic behavior via long-term potentiation and depression, the proposed flexible memristor device showed an outstanding accuracy of 96.44 % in image recognition tasks under extreme bending conditions, employing the MNIST and Fashion-MNIST datasets. The N:TaOx memristor deposited at low temperatures thus exhibits significant potential for use in wearable neuromorphic hardware applications due to its high density, high performance, reliability, and precise image recognition.

DenseHillNet: a lightweight CNN for accurate classification of natural images.

The detection of natural images, such as glaciers and mountains, holds practical applications in transportation automation and outdoor activities. Convolutional neural networks (CNNs) have been widely employed for image recognition and classification tasks. While previous studies have focused on fruits, land sliding, and medical images, there is a need for further research on the detection of natural images, particularly glaciers and mountains. To address the limitations of traditional CNNs, such as vanishing gradients and the need for many layers, the proposed work introduces a novel model called DenseHillNet. The model utilizes a DenseHillNet architecture, a type of CNN with densely connected layers, to accurately classify images as glaciers or mountains. The model contributes to the development of automation technologies in transportation and outdoor activities. The dataset used in this study comprises 3,096 images of each of the "glacier" and "mountain" categories. Rigorous methodology was employed for dataset preparation and model training, ensuring the validity of the results. A comparison with a previous work revealed that the proposed DenseHillNet model, trained on both glacier and mountain images, achieved higher accuracy (86%) compared to a CNN model that only utilized glacier images (72%). Researchers and graduate students are the audience of our article.

Robust two-dimensional PCANet with F-norm distance metric

The principal component analysis network (PCANet) is a simplified convolutional neural network (CNN) implemented using a deep subspace learning model. Since its typical PCA filters utilize the squared Frobenius-norm (F-norm) as the distance metric, it exhibits extreme sensitivity to outliers hidden within the data. Furthermore, PCANet faces challenges in extracting structural information in the row and column directions of images as the filters need to be pre-vectorized. To address these issues, we propose a robust bi-directional two-dimensional PCANet (RBDPCANet), wherein the filters utilize the F-norm instead of the squared F-norm to mitigate the impact of outliers. Additionally, the relationship between reconstruction errors and projection distances is simultaneously considered in the objective function. After a concise analytical description of the PCANet concept and related work, we present the methodology for generating three convolutional kernels in the form of RBDPCANet-1, RBDPCANet-2, and RBDPCANet-3 algorithms. These convolution kernels are iteratively obtained and are suitable for numerical implementation. Each of the three algorithms uses another optimization objective function. It has been shown that in image recognition tasks, the developed methods are characterized by better resistance to noise compared to the classic PCANet approach. The rotation invariance and convergence of RBDPCANet are demonstrated according to theoretical principles. The efficacy of the proposed method is validated using images from publicly available datasets. Extensive experimental results on these datasets demonstrate that the proposed RBDPCANet not only significantly boosts the efficiency of image classification but also has strong resistance to occlusions, changes in lighting, changes in orientation, etc. Our implementation code is available at https://github.com/Wesley-li1/RBDPCANet-.

MALARIA DETECTION USING CNN

The primary objective of this project is to develop a robust and efficient malaria detection system based on Convolutional Neural Networks (CNNs). CNNs have demonstrated remarkable performance in image recognition tasks, making them ideal for analyzing medical images such as blood smears for the presence of malaria parasites. By harnessing the power of CNN architecture, our goal is to create a highly accurate and reliable system capable of detecting malaria parasites with precision and consistency. Furthermore, we aim to ensure the efficiency of the developed system, allowing for the rapid processing of large volumes of blood smear images. This efficiency is crucial for facilitating timely diagnosis and treatment, particularly in regions where malaria is prevalent. By optimizing the computational processes involved in malaria detection, we seek to streamline the diagnostic workflow and enhance the overall effectiveness of malaria control efforts.In addition to accuracy and efficiency, adaptability is another key objective of this project. We aim to design a malaria detection system that can effectively handle variations in image quality and parasite morphology. This adaptability ensures that the system remains robust and reliable across different datasets and real-world scenarios, enabling healthcare professionals to confidently rely on its diagnostic capabilities.

Pre-rotation Only at Inference-Stage: A Way to Rotation Invariance of Convolutional Neural Networks

The popular convolutional neural networks (CNN) require data augmentation to achieve rotation invariance. We propose an alternative mechanism, Pre-Rotation Only at Inference stage (PROAI), to make CNN rotation invariant. The overall idea is to learn how the human brain observe images. At the training stage, PROAI trains a CNN with a small number using images only at one orientation. At the inference stage, PROAI introduces a pre-rotation operation to rotate each test image into its all-possible orientations and calculate classification scores using the trained CNN with a small number of parameters. The maximum of these classification scores is able to simultaneously estimate both the category and the orientation of each test image. The specific benefits of PROAI have been experimented on rotated image recognition tasks. The results shows that PROAI improves both the classification and orientation estimation performance while greatly reduced the numbers of parameters and the training time. Codes and datasets are publicly available at https://github.com/automlresearch/FRPRF.

Containerized deep learning in agriculture: Orchestrating GoogleNet with Kubernetes on high performance computing

SummarySmart Farming has become a cornerstone of modern agriculture, offering data‐driven insights and automation that optimize resource utilization and increase crop yields. The use of cutting‐edge technologies in agriculture has given rise to Smart Farming, which has transformed traditional farming practices into efficient, data‐driven operations. This paper explores the synergy between high‐performance computing (HPC) systems, Kubernetes orchestration, GoogleNet architecture, and containerization to redefine the future of farming. At the heart of this transformation lies the GoogleNet architecture, a deep learning powerhouse recognized for its efficiency and accuracy in image recognition tasks. The orchestration capabilities of Kubernetes, a versatile tool for managing containerized workloads efficiently on HPC clusters. Hence, in this work, we investigate the intricacies of deploying GoogleNet‐based deep learning models within containerized environments orchestrated by Kubernetes on HPC infrastructure. It explores resource optimization, scalability, security, and adaptability, all tailored to the unique demands of the agricultural domain to evaluate the effectiveness of the given technique it is compared with the existing techniques namely Hermes, Horus, CYBELE, and RZ‐SHAN. The attained ranges of proposed method of various measures of accuracy, precision, recall, and F1‐score are 98.65%, 97.45%, 97.87%, and 98.12% for the Pilot Wheat Ear dataset. Also, the processing time for the proposed approach is 181.50 and 120.2 m for the Pilot Wheat Ear Dataset and the Pilot soya bean farming dataset. The latency of the proposed approach attains a lower value of 1.5 and 1.1 s pilot soya bean farming dataset and Pilot Wheat Ear dataset. The experimental outcome demonstrates the efficiency of the proposed approaches to improve Smart Farming agriculture.

Improved stochastic configuration networks with vision patch fusion method for industrial image classification

This paper contributes to the advancement of stochastic configuration neural networks (SCN) in the field of visual applications. The proposed image classification randomized algorithm is an extension of deep stochastic configuration networks (DeepSCN), known as an improved SCN with vision patch fusion (Vi-SCN). Compared to existing two dimensional stochastic configuration networks (2DSCN) and DeepSCN, we have developed an incremental modeling method that employs a stochastic configuration patch fusion method. This method extracts randomly fused image features from three-channel high-resolution image data, improving the network's ability to extract features. Moreover, we have introduced a strategy for dynamically determining the depth structure of the network, enabling flexible adjustments to the network structure and the number of nodes in each layer based on the complexity of image recognition tasks. Through a series of comparisons on four image classification benchmark datasets, we assess the superior learning performance of our design compared to 2DSCN and DeepSCN. Furthermore, to evaluate the performance of Vi-SCN in practical visual application scenarios, we collect three industrial datasets with distinct characteristics. Vi-SCN demonstrate outstanding performance in all three tasks, showing its significant potential in image recognition.

Plant Leaf Disease Detection Using CNN

The existence of pests and diseases in plants and crops has a substantial impact on agricultural production within a country. Monitoring plants meticulously to detect and identify diseases is a common practice among farmers and experts. However, this approach is often laborious, costly, and not entirely reliable. To mitigate this issue, we propose a Disease Recognition Model based on leaf image classification. Our objective is to detect plant diseases using image processing techniques, specifically leveraging Convolutional Neural Networks (CNNs). CNNs are a category of artificial neural networks tailored for handling pixel-based inputs, particularly adept at image recognition tasks. Abbreviation- CNN (Convolutional Neural Network), SVM (Support Vector Machine), KNN (K Nearest Neighbour), ANN (Artificial Neural Network), GPU (Graphics processing unit)