Learning-based Solution Research Articles

Dog’s Breed Prediction using Deep Learning

Abstract: Dog Breed Prediction from Images using Deep Learning, in this project, we propose a Convolutional Neural Network (CNN) based approach for predicting dog breeds from images. With the increasing popularity of dogs as pets and the need for proper care and maintenance, it is essential to identify the breed of a dog accurately. However, manual identification can be time-consuming and prone to errors. Therefore, we propose a deep learning-based solution that can predict the breed of a dog with high accuracy. We train a CNN model on a large dataset of dog images, where each image is labeled with the breed of the dog. The model learns to extract features from the images and classify them into one of the 180 breeds. We evaluate the performance of the model using standard metrics and compare it with other machine learning algorithms. Our experimental results show that the proposed CNN model outperforms other algorithms in predicting dog breeds from images. The model achieves an accuracy of 90%, making it a reliable tool for dog breed prediction. Additionally, we develop a web application that allows users to upload an image of a dog and predict its breed using our model. This project demonstrates the effectiveness of deep learning techniques in predicting dog breeds from images. The proposed CNN model can be a valuable tool for veterinarians, pet owners, and breeders, helping them identify the breed of a dog accurately and efficiently.

An analysis of multi-agent reinforcement learning for decentralized inventory control systems

Most solutions to the inventory management problem assume a centralization of information that is incompatible with organizational constraints in supply chain networks. The problem can be naturally decomposed into sub-problems, each associated with an independent entity, turning it into a multi-agent system. A decentralized solution to inventory management using multi-agent reinforcement learning (MARL) is proposed where each entity is controlled by an agent. Three multi-agent variations of the proximal policy optimization algorithm are investigated through simulations of different supply chain networks and levels of uncertainty. A framework is deployed, which relies on offline centralization during simulation-based policy identification but enables decentralization when the policies are deployed online to the real system. Results show that reducing information sharing constraints in training enables MARL to perform comparatively to a centralized learning-based solution when deployed, and to outperform a distributed model-based solution in most cases, whilst respecting the information constraints of the system.

Flood Susceptibility Mapping for Kedah State, Malaysia: Geographics Information System-Based Machine Learning Approach

ABSTRACT Background: The world economy is significantly impacted by floods. Identifying flood risk is essential to flood mitigation techniques. Aim: The primary goal of this study is to create a geographic information system (GIS)-based flood susceptibility map for the study area. Methods: Ten flood-influencing factors from a geospatial database were taken into account when mapping the flood-prone areas. Every element demonstrated a robust relationship with the probability of flooding. Results: The highest contributing elements for the flood disaster in the study region were drainage density, distance, and the curvature. Flood susceptibility models’ performance was validated using standard statistical measures and AUC. The ROC curves demonstrated that all ensemble models had good performance on the validation data sets (AUC = >0.97) with high accuracy scores of 0.80. Based on the flood susceptibility maps, most of the northwest regions of the study area are more likely to flood because of low land areas, areas with a lower gradient slope, linear and concave shape curvature, high drainage density with high rainfall, more “water bodies,” “crops land,” and “built areas,” abundance on sea and surface water, and Quaternary types of soil feature and so on. The very high flood susceptibility class accounts for 18.2% of the study area, according to the RF-embedding model, whereas the high, moderate, low, and very low susceptibility classes were found at about 20.0%, 24.6%, 24.3%, and 12.9%, respectively. Conclusion: In comparison with other commonly used applied approaches, this research presents a novel modeling approach for flood susceptibility that integrates machine learning and geospatial data. It has been found to be stronger and more efficient, highly accurate, has good prediction performance, and is less biased. Overall, our research into machine learning-based solutions points in a positive path technologically and can serve as a reference manual for future research and applications for academic specialists and decision-makers.

The Quest for the Application of Artificial Intelligence to Whole Slide Imaging: Unique Prospective from New Advanced Tools

The introduction of machine learning in digital pathology has deeply impacted the field, especially with the advent of whole slide image (WSI) analysis. In this review, we tried to elucidate the role of machine learning algorithms in diagnostic precision, efficiency, and the reproducibility of the results. First, we discuss some of the most used tools, including QuPath, HistoQC, and HistomicsTK, and provide an updated overview of machine learning approaches and their application in pathology. Later, we report how these tools may simplify the automation of WSI analyses, also reducing manual workload and inter-observer variability. A novel aspect of this review is its focus on open-source tools, presented in a way that may help the adoption process for pathologists. Furthermore, we highlight the major benefits of these technologies, with the aim of making this review a practical guide for clinicians seeking to implement machine learning-based solutions in their specific workflows. Moreover, this review also emphasizes some crucial limitations related to data quality and the interpretability of the models, giving insight into future directions for research. Overall, this work tries to bridge the gap between the more recent technological progress in computer science and traditional clinical practice, supporting a broader, yet smooth, adoption of machine learning approaches in digital pathology.

Adaptive bridge model for compressed domain point cloud classification

The recent adoption of deep learning-based models for the processing and coding of multimedia signals has brought noticeable gains in performance, which have established deep learning-based solutions as the uncontested state-of-the-art both for computer vision tasks, targeting machine consumption, as well as, more recently, coding applications, targeting human visualization. Traditionally, applications requiring both coding and computer vision processing require first decoding the bitstream and then applying the computer vision methods to the decompressed multimedia signals. However, the adoption of deep learning-based solutions enables the use of compressed domain computer vision processing, with gains in performance and computational complexity over the decompressed domain approach. For point clouds (PCs), these gains have been demonstrated in the single available compressed domain computer vision processing solution, named Compressed Domain PC Classifier, which processes JPEG Pleno PC coding (PCC) compressed streams using a PC classifier largely compatible with the state-of-the-art spatial domain PointGrid classifier. However, the available Compressed Domain PC Classifier presents strong limitations by imposing a single, specific input size which is associated to specific JPEG Pleno PCC configurations; this limits the compression performance as these configurations are not ideal for all PCs due to their different characteristics, notably density. To overcome these limitations, this paper proposes the first Adaptive Compressed Domain PC Classifier solution which includes a novel adaptive bridge model that allows to process the JPEG Pleno PCC encoded bit streams using different coding configurations, now maximizing the compression efficiency. Experimental results show that the novel Adaptive Compressed Domain PC Classifier allows JPEG PCC to achieve better compression performance by not imposing a single, specific coding configuration for all PCs, regardless of its different characteristics. Moreover, the added adaptability power can achieve slightly better PC classification performance than the previous Compressed Domain PC Classifier and largely better PC classification performance (and lower number of weights) than the PointGrid PC classifier working in the decompressed domain.

Multimodal fine-grained grocery product recognition using image and OCR text

Automatic recognition of grocery products can be used to improve customer flow at checkouts and reduce labor costs and store losses. Product recognition is, however, a challenging task for machine learning-based solutions due to the large number of products and their variations in appearance. In this work, we tackle the challenge of fine-grained product recognition by first extracting a large dataset from a grocery store containing products that are only differentiable by subtle details. Then, we propose a multimodal product recognition approach that uses product images with extracted OCR text from packages to improve fine-grained recognition of grocery products. We evaluate several image and text models separately and then combine them using different multimodal models of varying complexities. The results show that image and textual information complement each other in multimodal models and enable a classifier with greater recognition performance than unimodal models, especially when the number of training samples is limited. Therefore, this approach is suitable for many different scenarios in which product recognition is used to further improve recognition performance. The dataset can be found at https://github.com/Tubbias/finegrainocr.

Estimation of flea beetle damage in the field using a multistage deep learning-based solution

Estimation of damage in plants is a key issue for crop protection. Currently, experts in the field manually assess the plots. This is a time-consuming task that can be automated thanks to the latest technology in computer vision (CV). The use of image-based systems and recently deep learning-based systems have provided good results in several agricultural applications. These image-based applications outperform expert evaluation in controlled environments, and now they are being progressively included in non-controlled field applications.A novel solution based on deep learning techniques in combination with image processing methods is proposed to tackle the estimate of plant damage in the field. The proposed solution is a two-stage algorithm. In a first stage, the single plants in the plots are detected by an object detection YOLO based model. Then a regression model is applied to estimate the damage of each individual plant. The solution has been developed and validated in oilseed rape plants to estimate the damage caused by flea beetle.The crop detection model achieves a mean precision average of 91% with a mAP@0.50 of 0.99 and a mAP@0.95 of 0.91 for oilseed rape specifically. The regression model to estimate up to 60% of damage degree in single plants achieves a MAE of 7.11, and R2 of 0.46 in comparison with manual evaluations done plant by plant by experts. Models are deployed in a docker, and with a REST API communication protocol they can be inferred directly for images acquired in the field from a mobile device.

Attention-based Imputation of Missing Values in Electronic Health Records Tabular Data.

The imputation of missing values (IMV) in electronic health records tabular data is crucial to enable machine learning for patient-specific predictive modeling. While IMV methods are developed in biostatistics and recently in machine learning, deep learning-based solutions have shown limited success in learning tabular data. This paper proposes a novel attention-based missing value imputation framework that learns to reconstruct data with missing values leveraging between-feature (self-attention) or between-sample attentions. We adopt data manipulation methods used in contrastive learning to improve the generalization of the trained imputation model. The proposed self-attention imputation method outperforms state-of-the-art statistical and machine learning-based (decision-tree) imputation methods, reducing the normalized root mean squared error by 18.4% to 74.7% on five tabular data sets and 52.6% to 82.6% on two electronic health records data sets. The proposed attention-based missing value imputation method shows superior performance across a wide range of missingness (10% to 50%) when the values are missing completely at random.

A deep learning-based surrogate model for probabilistic analysis of high-speed railway tunnel crown settlement in spatially variable soil considering construction process

In reality, many high-speed railway tunnels are built by sequential excavation method in spatially variable soil. This study introduces an efficient deep learning-based solution for the probabilistic analysis of tunnel crown settlements considering the spatial variability of surrounding soil. The two-dimensional convolutional neural network (2D-CNN) is employed to uncover the implicit correlation between the soil elastic modulus random field and tunnel crown settlements. The correlation coefficient (R) of the surrogate model is greater than 0.9664. Meanwhile, the mean relative error of the predictions remains within 2.38%, under various scales of fluctuation (SOFs). An additional 10,000 samples were produced to assess the practical performance of the trained model in probabilistic analysis. The relative errors of predictions generally fall within 5%, with a minimum confidence interval of 98.15%. These results demonstrate that the developed 2D-CNN model can effectively substitute the traditional method in predicting tunnel crown settlements considering soil spatial variability.

An Integrated Approach for Amazon Electronic Products Reviews by Using Sentiment Analysis

In our routine life, we interact a wide range of products, and frequently browse through digital media platforms to access their quality. Although the accessibility of online platforms, consumers often find it challenging to swiftly judge the quality of products on the basis of customer reviews. To cope this situation, the study addresses this problem by suggesting a machine learning-based solution to categorize product reviews. For this, we employ various machine learning techniques, including Random Forest, Naïve Bayes, Support Vector Machine (SVM), Stochastic Gradient Descent (SGD) Classifier, and Bidirectional Encoder Representations from Transformers (BERT). In our model, we incorporate pre-processing methods for prepare the dataset for training and utilize feature extraction techniques such as TF-IDF and word2vec which are then applied to different classifiers to analyze the reviews. Moreover, we conduct this study by using the Amazon Electronics category dataset, it reveals that BERT outperforms other classifiers with a performance score of 0.8896. Therefore, this technique not only streamlines the procedure of evaluating product quality but also enhances the accuracy of review classification, giving a real-world solution for consumers and businesses alike.

Suitable and Style-Consistent Multi-Texture Recommendation for Cartoon Illustrations

Texture plays an important role in cartoon illustrations to display object materials and enrich visual experiences. Unfortunately, manually designing and drawing an appropriate texture is not easy even for proficient artists, let alone novice or amateur people. While there exist tons of textures on the Internet, it is not easy to pick an appropriate one using traditional text-based search engines. Although several texture pickers have been proposed, they still require the users to browse the textures by themselves, which is still labor-intensive and time-consuming. In this article, an automatic texture recommendation system is proposed for recommending multiple textures to replace a set of user-specified regions in a cartoon illustration with visually pleasant look. Two measurements, the suitability measurement and the style-consistency measurement, are proposed to make sure that the recommended textures are suitable for cartoon illustration and at the same time mutually consistent in style. The suitability is measured based on the synthesizability, cartoonity, and region fitness of textures. The style-consistency is predicted using a learning-based solution since it is subjective to judge whether two textures are consistent in style. An optimization problem is formulated and solved via the genetic algorithm. Our method is validated on various cartoon illustrations, and convincing results are obtained.

A novel multi-task machine learning classifier for rare disease patterning using cardiac strain imaging data

To provide accurate predictions, current machine learning-based solutions require large, manually labeled training datasets. We implement persistent homology (PH), a topological tool for studying the pattern of data, to analyze echocardiography-based strain data and differentiate between rare diseases like constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM). Patient population (retrospectively registered) included those presenting with heart failure due to CP (n = 51), RCM (n = 47), and patients without heart failure symptoms (n = 53). Longitudinal, radial, and circumferential strains/strain rates for left ventricular segments were processed into topological feature vectors using Machine learning PH workflow. In differentiating CP and RCM, the PH workflow model had a ROC AUC of 0.94 (Sensitivity = 92%, Specificity = 81%), compared with the GLS model AUC of 0.69 (Sensitivity = 65%, Specificity = 66%). In differentiating between all three conditions, the PH workflow model had an AUC of 0.83 (Sensitivity = 68%, Specificity = 84%), compared with the GLS model AUC of 0.68 (Sensitivity = 52% and Specificity = 76%). By employing persistent homology to differentiate the “pattern” of cardiac deformations, our machine-learning approach provides reasonable accuracy when evaluating small datasets and aids in understanding and visualizing patterns of cardiac imaging data in clinically challenging disease states.

Extensive data engineering to the rescue: building a multi-species katydid detector from unbalanced, atypical training datasets.

Passive acoustic monitoring (PAM) is a powerful tool for studying ecosystems. However, its effective application in tropical environments, particularly for insects, poses distinct challenges. Neotropical katydids produce complex species-specific calls, spanning mere milliseconds to seconds and spread across broad audible and ultrasonic frequencies. However, subtle differences in inter-pulse intervals or central frequencies are often the only discriminatory traits. These extremities, coupled with low source levels and susceptibility to masking by ambient noise, challenge species identification in PAM recordings. This study aimed to develop a deep learning-based solution to automate the recognition of 31 katydid species of interest in a biodiverse Panamanian forest with over 80 katydid species. Besides the innate challenges, our efforts were also encumbered by a limited and imbalanced initial training dataset comprising domain-mismatched recordings. To overcome these, we applied rigorous data engineering, improving input variance through controlled playback re-recordings and by employing physics-based data augmentation techniques, and tuning signal-processing, model and training parameters to produce a custom well-fit solution. Methods developed here are incorporated into Koogu, an open-source Python-based toolbox for developing deep learning-based bioacoustic analysis solutions. The parametric implementations offer a valuable resource, enhancing the capabilities of PAM for studying insects in tropical ecosystems. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'.

Enhancing the Safety of Autonomous Vehicles in Adverse Weather by Deep Learning-Based Object Detection

Recognizing and categorizing items in weather-adverse environments poses significant challenges for autonomous vehicles. To improve the robustness of object-detection systems, this paper introduces an innovative approach for detecting objects at different levels by leveraging sensors and deep learning-based solutions within a traffic circle. The suggested approach improves the effectiveness of single-stage object detectors, aiming to advance the performance in perceiving autonomous racing environments and minimizing instances of false detection and low recognition rates. The improved framework is based on the one-stage object-detection model, incorporating multiple lightweight backbones. Additionally, attention mechanisms are integrated to refine the object-detection process further. Our proposed model demonstrates superior performance compared to the state-of-the-art method on the DAWN dataset, achieving a mean average precision (mAP) of 99.1%, surpassing the previous result of 84.7%.

Sim-to-Real Application of Reinforcement Learning Agents for Autonomous, Real Vehicle Drifting

Enhancing the safety of passengers by venturing beyond the limits of a human driver is one of the main ideas behind autonomous vehicles. While drifting is mostly witnessed in motorsports as an advanced driving technique, it could provide many possibilities for improving traffic safety by avoiding accidents in extreme traffic situations. The purpose of the research presented in this article is to provide a machine learning-based solution to autonomous drifting as a proof of concept for vehicle control at the limits of handling. To achieve this, reinforcement learning (RL) agents were trained for the task in a MATLAB/Simulink-based simulation environment, using the state-of-the-art Soft Actor–Critic (SAC) algorithm. The trained agents were tested in reality at the ZalaZONE proving ground on a series production sports car with zero-shot transfer. Based on the test results, the simulation environment was improved through domain randomization, until the agent could perform the task both in simulation and in reality on a real test car.

RL-TEE: Autonomous Probe Guidance for Transesophageal Echocardiography Based on Attention-Augmented Deep Reinforcement Learning

Ultrasound image acquisition in conventional transesophageal echocardiography (TEE) requires complex manual operation of the probe in the esophagus based on the interpretation of ultrasound images and in-depth knowledge of the cardiac anatomy. In this work, we formulate the TEE probe guidance task as a reinforcement learning (RL) problem, and present the first learning-based solution to 3-DOF control of a TEE probe based on the ultrasound image feedback, named RL-TEE, in order to mimic the visual search and navigation strategies of expert echocardiographers. The probe-tissue interaction in TEE is carefully modeled in our framework by considering both the requirements for navigation towards the standard views and compliance in the esophageal environment. Furthermore, we propose a hybrid deep Q-network model that augments a convolutional neural network backbone with self-attention mechanisms to better capture spatial information in ultrasound images to guide navigation decisions. The presented methods are preliminarily validated in a TEE simulation environment built with data from 25 subjects to acquire four standard views of the heart. Our results show that the proposed method can effectively learn to accurately and compliantly guide the probe movement for TEE standard view acquisition tasks and has a good generalization ability to unseen patient data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The motivation of this paper is to realize 3-DOF movement guidance of a TEE probe to acquire the standard views of the heart based on the real-time images, which can be applied to existing robotic control systems or used to assist novice echocardiographers in TEE examination, thereby relieving operator workload and improving ease of use. This paper suggests a novel approach that uses the deep RL technique to achieve automatic interpretation of TEE images and intelligent guidance of the probe movement. The RL framework is designed to take into account both the navigation efficiency and compliance with the esophageal environment for the targeted intracorporeal application. A hybrid deep Q-network model that augments a convolutional neural network with attention mechanisms is designed to better capture spatial information from ultrasound images to predict the probe movement. The effectiveness of the framework is preliminarily validated in extensive experiments in a simulation environment built with real patient data. The proposed method can be applied in clinical use to provide real-time TEE probe guidance for novice echocardiographers, and can be integrated with a robotic system to fully automate the TEE acquisition, thereby relieving the doctors from tedious manual operation to focus on the diagnosis and treatment.

How to detect occluded crosswalks in overview images? Comparing three methods in a heavily occluded area

Crosswalk presence data is crucial for pedestrian safety and urban planning. However, obtaining such data at a large scale is often challenging due to the high cost associated with traditional collection methods. While automated methods based on computer vision have been explored to detect crosswalks from aerial images, a major obstacle to their application is the handling of candidate crosswalks occluded by objects or shadows in the aerial imagery. To address this challenge, this study explores different deep learning-based solutions, including the aerial-view and street-view methods, which are commonly used, and a combination of the two − dual-perspective method. Deep learning models based on Convolutional Neural Networks with VGG16 architecture were trained using 16,815 images to automatically detect crosswalks from both aerial and street view images. To compare the performance of these methods in handling occlusions, 1,378 images from a heavily occluded area were processed separately by the three methods. The results showed that the aerial-view method suffered the most when dealing with images from a heavily occluded area, resulting in the lowest accuracy, precision, recall, and F1 score among the three methods. On the other hand, the street-view method outperformed the aerial-view method significantly. The dual-perspective method demonstrated the highest accuracy and precision values, indicating its superiority in accurately predicting the location of a crosswalk. However, the street-view method exhibited the highest recall value, highlighting its superior ability to recover an occluded crosswalk among all methods.

Blockchain-Enabled Intelligent IoT Protocol for High-Performance and Secured Big Financial Data Transaction

In the recent era, the communication network with the support of many wireless technologies is giving benefits for remote access. Such a communication model increases the flexibility for data storage with the management of network resources efficiently with the integration of the Internet of Things (IoT). Although, the industrial Internet of Things (IIoT) has enabled the development of numerous machine learning-based solutions to provide real-time applications. However, most of the solutions are not prepared to cope with heterogeneous services and the huge amount of device data in the digital world. Furthermore, conducting financial transactions over the Internet raises several security issues. Such restrictions compromise the sensitive data of financial institutions and also degrade the trust of network users in the system. Thus, this article presents a secured blockchain model for high-performance computing in a big data environment, which aims to protect the business activities for financial interaction with intelligent services of software-defined network (SDN) architecture. First, to keep the security credentials, the SDN controller creates an association between the IoT devices and maintains local and global records. Second, the machine learning approach is explored using reliable and fault-tolerant methods to support the network scalability and extract the updated routing information for transmitting financial data. The proposed protocol also provides data integrity with a high level of network availability and copes with the financial security of big data by investigating cryptographic approaches. Our proposed protocol is tested using simulation, and various experiments are performed to show its efficacy in terms of network throughput, computing overhead, data delay, response time, and dropped packets as compared to tunicate swarm algorithm-based optimized routing mechanism (TORM) and RouteChain.

Towards Automated RISC-V Microarchitecture Design with Reinforcement Learning

Microarchitecture determines the implementation of a microprocessor. Designing a microarchitecture to achieve better performance, power, and area (PPA) trade-off has been increasingly difficult. Previous data-driven methodologies hold inappropriate assumptions and lack more tightly coupling with expert knowledge. This paper proposes a novel reinforcement learning-based (RL) solution that addresses these limitations. With the integration of microarchitecture scaling graph, PPA preference space embedding, and proposed lightweight environment in RL, experiments using commercial electronic design automation (EDA) tools show that our method achieves an average PPA trade-off improvement of 16.03% than previous state-of-the-art approaches with 4.07× higher efficiency. The solution qualities outperform human implementations by at most 2.03× in the PPA trade-off.

Referee-Meta-Learning for Fast Adaptation of Locational Fairness

When dealing with data from distinct locations, machine learning algorithms tend to demonstrate an implicit preference of some locations over the others, which constitutes biases that sabotage the spatial fairness of the algorithm. This unfairness can easily introduce biases in subsequent decision-making given broad adoptions of learning-based solutions in practice. However, locational biases in AI are largely understudied. To mitigate biases over locations, we propose a locational meta-referee (Meta-Ref) to oversee the few-shot meta-training and meta-testing of a deep neural network. Meta-Ref dynamically adjusts the learning rates for training samples of given locations to advocate a fair performance across locations, through an explicit consideration of locational biases and the characteristics of input data. We present a three-phase training framework to learn both a meta-learning-based predictor and an integrated Meta-Ref that governs the fairness of the model. Once trained with a distribution of spatial tasks, Meta-Ref is applied to samples from new spatial tasks (i.e., regions outside the training area) to promote fairness during the fine-tune step. We carried out experiments with two case studies on crop monitoring and transportation safety, which show Meta-Ref can improve locational fairness while keeping the overall prediction quality at a similar level.