Cloud APIs Research Articles

CCeACF: content and complementarity enhanced attentional collaborative filtering for cloud API recommendation

CCeACF: content and complementarity enhanced attentional collaborative filtering for cloud API recommendation

Health of Things Melanoma Detection System—detection and segmentation of melanoma in dermoscopic images applied to edge computing using deep learning and fine-tuning models

According to the World Health Organization (WHO), melanoma is a type of cancer that affects people globally in different parts of the human body, leading to deaths of thousands of people every year worldwide. Intelligent diagnostic tools through automatic detection in medical images are extremely effective in aiding medical diagnosis. Computer-aided diagnosis (CAD) systems are of utmost importance for image-based pre-diagnosis, and the use of artificial intelligence–based tools for monitoring, detection, and segmentation of the pathological region are increasingly used in integrated smart solutions within smart city systems through cloud data processing with the use of edge computing. This study proposes a new approach capable of integrating into computational monitoring and medical diagnostic assistance systems called Health of Things Melanoma Detection System (HTMDS). The method presents a deep learning–based approach using the YOLOv8 network for melanoma detection in dermatoscopic images. The study proposes a workflow through communication between the mobile device, which extracts captured images from the dermatoscopic device and uploads them to the cloud API, and a new approach using deep learning and different fine-tuning models for melanoma detection and segmentation of the region of interest, along with the cloud communication structure and comparison with methods found in the state of the art, addressing local processing. The new approach achieved satisfactory results with over 98% accuracy for detection and over 99% accuracy for skin cancer segmentation, surpassing various state-of-the-art works in different methods, such as manual, semi-automatic, and automatic approaches. The new approach demonstrates effective results in the performance of different intelligent automatic models with real-time processing, which can be used in affiliated institutions or offices in smart cities for population use and medical diagnosis purposes.

The Facial Recognition Technology in Academic Attendance: A Comparative Study for Real-Time Management

In today’s academic settings, managing daily tasks like attendance tracking has become increasingly burdensome. Traditional methods of manual attendance taking are inefficient and time-consuming, particularly with growing numbers of students and staff. To address this challenge, various approaches, including face identification systems, have been developed. This research introduces a smart face recognition (FR) system for managing attendance efficiently. The system employs multiple face recognition methodologies such as Local Binary Histogram, PCA/Eigen Face Recognizer, and Fisher Face Recognizer, enhancing system performance. These methods are combined using Ensemble Fusion to improve accuracy. Additionally, the system utilizes Multitask Cascaded Convolutional Network for face detection and attribute extraction. Extracted attributes are matched with stored facial templates to identify recognized faces and mark attendance. Integration with Cloud API facilitates record-keeping. The system also includes a feedback and notification system for process status indication. Results indicate that the proposed system achieves 82.1% accuracy in face recognition and requires minimal time (0.000081s) to predict and mark attendance.

Enhancing Campus Connectivity: A Smart Intra-Transit Strategy for Efficient Vehicle Throughput

An advanced vehicle monitoring, and seat availability system is designed to monitor the vehicles from any source location A to destination location B in real time to the passengers. The proposed system would make good use of modern technologies via leveraging ultrasonic sensors, cloud API for example Thing Speak, and in-house Word Press software. The system track’s location, speed, and passenger count in real-time. This data fuels optimized scheduling and route planning, maximizing seat occupancy, and boosting overall productivity. A user-friendly dashboard visualizes vehicle activity within designated time slots, empowering faculty, and administrators with data-driven insights for improved resource allocation and scheduling. By comprehensively monitoring location, speed, and passenger count, the system ensures efficient Electric Vehicle operation within the campus confines, ultimately revolutionizing campus transportation and maximizing resource utilization.

Attentional Feature Erase: Towards task-wise transferable adversarial attack on cloud vision APIs

Recent works have shown that adversarial examples (AEs) can attack and successfully transfer across various neural networks, highlighting the potential danger they pose. However, current approaches that focus on task-specific loss functions may not be as effective across different tasks. Additionally, the use of cloud APIs in practice, which often involve combining multiple tasks, also weakens the effectiveness of existing attacks. To address these issues, we propose a method called Attentional Feature Erase, which is a task-agnostic attack with improved cross-task transferability and effectiveness on computer vision-based cloud APIs. We view the transferability of AEs as a latent contribution for each layer of deep neural networks. By focusing on the intermediate layers of model backbones and reducing high-value features in each intermediate feature map, we are able to maximize the attack performance. Additionally, to better aggregate the gradients and generate adversarial perturbations during backward propagation, Transferability Regularizer is proposed to calculate the attention heatmap for each intermediate feature map and systematically combine the gradients. Comprehensive set of experiments on the Google Cloud Vision APIs and public available datasets (i.e. ImageNet, PASCAL VOC and MS COCO) show that the proposed AFE attack is more effective and has better transferability compared to the state-of-the-art baselines.

The Mechanism of Identification in the External Dissemination of the Discourse of “Community of a Shared Future for Mankind”

The “Community of a Shared Future for Mankind” has emerged as a pivotal notion in contemporary international relations and global governance. Drawing from President Xi Jinping’s public speeches concerning the topic, this article explores the rhetorical strategies by using Kenneth Burke’s Identification Theory. With the help of word cloud API and the Doccano text annotation tool, the research conducts a systematic analysis to delve into the application and tangible impacts of the identification mechanism in the external dissemination of the discourse. It is founded that the strategies of identification by sympathy and inaccuracy are predominant in the discourse, with identification by antithesis as a supplementary method.

Poisoning QoS-aware cloud API recommender system with generative adversarial network attack

With the proliferation and deepening of service-oriented architecture, more and more enterprises and organizations are exposing their computing functions and big data to the Internet in the form of cloud APIs to support service-oriented software development. This has resulted in a plethora of cloud APIs with similar functionality appearing on the Web, drowning users in a sea of cloud API choices. To solve this problem, quality of service (QoS)-aware recommender system is then widely applied to the selection of cloud APIs. Due to the dynamic and open network environment, the QoS-aware cloud API recommender systems are vulnerable to data poisoning attacks, where attackers inject poisoned data to skew the recommender system and make the recommendation direction follow the attacker's will. Given the lack of data poisoning attack methods and robustness analysis for existing QoS-aware cloud API recommender systems, in this work, we first built a general poisoning attack framework for QoS-aware cloud API recommender systems to elucidate and standardize the attack process. Then, we proposed a deep learning-based poison attack approach, which uses generative adversarial network (GAN) to learn the cloud API QoS data distribution of real users in an adversarial way, so as to generate high-quality fake user attack vectors. We conducted extensive experiments on real-world QoS datasets, and the experimental results show that our proposed GAN-based poisoning attack is effective and can better hide itself from being detected. In addition, we analyzed the data poisoning attack mechanism and the robustness of the cloud API recommender system based on four categories of twelve recommendation methods, thereby raising awareness about the security of cloud API recommendation and helping the recommender system defenders to develop more targeted defense strategies.

Design and Implementation of Raspberry Pi Face Recognition + Thermal Imaging Temperature Measurement and Body Temperature Data Cloud Processing System

For temperature measurement in public places, this paper proposes and realizes a Raspberry Pi face recognition + thermal imaging temperature measurement and temperature data cloud processing system. This system uses a Raspberry Pi device as the temperature measurement terminal, connects the MLX90640 thermal imaging sensor and the BME280 environmental sensor for data collection, and calculates the core temperature of the human body from the collected data. The temperature measurement terminal uses Python language with OpenCV library for image processing, Alibaba Cloud API for face recognition, and the server side uses Python language with Django library for development. The system can quickly measure temperature and count the temperature measurement list, and view it in the cloud in real time to improve management efficiency.

A novel watermarking framework for intellectual property protection of NLG APIs

Natural language generation (NLG) models have attracted extensive attention and applications due to their combination with powerful deep learning techniques. Many NLG models are encapsulated in cloud APIs serving commercial organizations, which have become very important profitable services for these institutions. However, cloud platforms may suffer from model extraction attacks that aim to imitate the functionality of these NLG models in practical applications, thus infringing the intellectual property (IP) of the NLG APIs. Unfortunately, most current watermarking methods for protecting deep model IP are not directly applicable to IP protection of NLG APIs. In addition, the semantic similarity between the watermarked texts generated by the baseline method and the original texts is not high enough, which can be easily detected by attackers. To make up these gaps, we propose a novel watermarking framework which embeds watermarks by conducting lexical modification to the outputs of the NLG models, and uses the corresponding watermark identification method can identify the attackers and protect the IP of NLG APIs. Experiment result shows that our proposed watermarking method not only generates watermarked texts with higher semantically similar to the original texts but also achieves better identifiable performance compared with the baseline method. In addition, our watermarking method also exhibits outstanding performance in other aspects such as transferability, watermark undetectability and robustness.

Recognizing speech in voice messages

The level of development of information technology makes it possible to use speech recognition technologies in a wide range of human life and activities. It is very convenient to use the voice interface: voice search for the necessary documents, dialing a phone number, managing IOT devices, voice navigation, simple text dictation. Since the natural language interface provides an additional convenience for a person when typing, sending voice messages has become common among users. In this case, voice messages are audio files. But it is not always available and convenient for the recipient to listen to such messages. This problem can be solved with the help of an automatic speech recognition system (ASR). The article describes the stages and elements of the process of processing and recognition of natural language by audio signal. Modern technologies of automatic speech recognition and problems with choosing among them are indicated. Modern automatic speech recognition (ASR) systems understand fully spontaneous speech that is natural, not memorized, contains signs of stuttering or even minor errors. At the same time, they are still too expensive to develop from scratch. So companies are faced with a choice between using the cloud API for ASR developed by the tech giants and using open source solutions. The analysis of the latest research and publications on the processing of voice data is considered. A software solution for automatic conversion of voice messages into text is proposed. The interface to the voice signal delivery system is proposed to be made as a chat bot in the messenger. The article presents the main components of the system, the algorithm of the chat bot, modern technologies for the development, implementation and configuration of the chat bot in the messenger

A Long-Range Internet of Things-Based Advanced Vehicle Pollution Monitoring System with Node Authentication and Blockchain

According to United Nations (UN) 2030 agenda, the pollution detection system needs to be improved for the establishment of fresh air to obtain healthy life of living things. There are many reasons for the pollution and one of the reasons for pollution is from the emissions of the vehicles. Currently digital technologies such as the Internet of Things and Long-Range are showing significant impact on establishment of smart infrastructure for achieving the sustainability. Based on this motivation, this study implemented a sensor node and gateway-based Internet of Things architecture to monitor the air quality index value from any location through Long-Range communication, and Internet connectivity. To realize the proposed system, a customization of hardware is carried out and implemented the customized hardware i.e., sensor node and gateway in real-time. The sensor node is powered with node mapping to minimize the data redundancy. In this study, the evaluation metrics such as bit rate, receiver sensitivity, and time on air are evaluated by spreading factor (SF), code rate (CR), bandwidth, number of packets, payload size, preamble, and noise figure. The real-time sensor values are logged on the cloud server through sensor node and gateway. The sensor values recorded in the cloud server is compared with optimal values and concluded that the PM10, PM2.5 are high in the air and remaining values of NO2, O3, CO are optimal in the air. Along with this an architecture is proposed for interfacing the hardware with blockchain network through cloud server and API for node authentication.

Protecting Intellectual Property of Language Generation APIs with Lexical Watermark

Nowadays, due to the breakthrough in natural language generation (NLG), including machine translation, document summarization, image captioning, etc NLG models have been encapsulated in cloud APIs to serve over half a billion people worldwide and process over one hundred billion word generations per day. Thus, NLG APIs have already become essential profitable services in many commercial companies. Due to the substantial financial and intellectual investments, service providers adopt a pay-as-you-use policy to promote sustainable market growth. However, recent works have shown that cloud platforms suffer from financial losses imposed by model extraction attacks, which aim to imitate the functionality and utility of the victim services, thus violating the intellectual property (IP) of cloud APIs. This work targets at protecting IP of NLG APIs by identifying the attackers who have utilized watermarked responses from the victim NLG APIs. However, most existing watermarking techniques are not directly amenable for IP protection of NLG APIs. To bridge this gap, we first present a novel watermarking method for text generation APIs by conducting lexical modification to the original outputs. Compared with the competitive baselines, our watermark approach achieves better identifiable performance in terms of p-value, with fewer semantic losses. In addition, our watermarks are more understandable and intuitive to humans than the baselines. Finally, the empirical studies show our approach is also applicable to queries from different domains, and is effective on the attacker trained on a mixture of the corpus which includes less than 10% watermarked samples.

Construction of Water Conservancy and Energy Engineering Structure Platform Based on Cloud Computing

In view of the current water conservancy engineering system in solving the problem of water conservancy spatial information sharing and repeated investment in the construction of GIS systems by various water conservancy business departments, the research on the construction of water conservancy engineering structure platform based on cloud computing is proposed. We discussed the key technologies to realize the platform: pooled services, elastic cloud services, and spatial information services. The aggregation pooling service mainly uses the pooling mechanism and improves the responsiveness of platform services by creating object instances in advance. Based on the Cloud API provided by the cloud computing platform (such as creating a virtual machine, starting a virtual machine at runtime, and obtaining the real-time status of a virtual machine) and the REST Admin API provided by the water conservancy basic service module, using cloud computing technology ideas, we designed and implemented elastic cloud service technology. Enable the water conservancy spatial information service platform to be deployed in the cloud computing environment, support cloud architecture, and provide flexible computing capabilities, because the three types of services have different requirements for CPU and network utilization, the weight coefficient can be assigned to them. The CPU slice map service is 0.1, the spatial query service is 0.3, and the spatial analysis processing service is 0.6. The network bandwidth slice map service is 0.5, the spatial query service is 0.3, and the spatial analysis processing service is 0.2. Reduce the waste of infrastructure resources; meet the concurrent access needs of a large number of users. Spatial information service aggregation is one of the core service capabilities of the platform; it can effectively aggregate multisource spatial information, generate new services, and provide new information.

An Augmented Reality Framework for Visualization of Internet of Things Data for Process Supervision in Factory Shop-Floor

Modern industrial environments provide a high level of digitalization, enabling great volumes of information to be used for shopfloor monitoring and control purposes. In cases where human operators participate in the production process, it is important to support operator’s intervention with Internet of Things (IoT) data and empower process supervision with constant information over Key Performance Indicators (KPIs). Towards this end, the proposed framework exploits Augmented Reality (AR) for the provision of an IoT data visualization solution to support KPIs based process supervision in the smart factory environment. The solution utilizes IoT-enabled components, in order to receive dynamic information over the status of sensors and resources spread across the factory shop-floor. The AR application facilitates the need for visualization of information available on the network, on the actual shopfloor components, which offer transparency and ease the supervision of production processes. The development of the AR and cloud API has been based on CSharp and Java, in a reusable framework that can be reused for similar industrial cases. The reconfigurability and reuse of the AR solution is supported by a template configuration, allowing industry practitioners to configure application features and external connections without any additional software development. The solution was validated on a thermoforming process supervision case of a white goods production system. The main goal was to use the AR framework to support operators during the process parameters design phase, and dynamic supervision of the shopfloor environment. It is demonstrated that the proposed system can be used for KPIs monitoring of the production cycles and for empowering process supervision and optimization.

Design Of A Secure Virtual File Storage System On Cloud Using Hybrid Cryptography

As Security is becoming more and more useful in the field of computing, users would like to be sure of how secure their files are on a system, as security is one of the most crucial fields in networking and file storage. Dependable file storage and access establish several security issues in a cloud computing. This research designed and implemented virtual secure file storage system on cloud using hybrid cryptography. The cryptography method used for file encryption and decryption is AES and SHA-2 hash function. It is implemented using Cloud APIs with REST calls or client libraries in PHP. The system interfaces were developed using HTML, CSS and JAVASCRIPT. Back end development was done using PHP, MYSQL and GCP Cloud Storage Library then the file encryption and decryption was achieved through PHP classes which includes open_ssl_file_encryption and decryption (AES) and also MCRYPT function. The proposed virtual system is also compared with some latest related works.

Self-Alignment for Black-Box Domain Adaptation of Image Classification

Recently, black-box domain adaptation attracts a lot of attention, which is a new concept to realize domain adaptation with only a cloud API service instead of the source data or well-trained source model, reflecting the focus on development of cloud services and concerns about data security. However, the existing black-box domain adaptation methods always only use high-confidence samples which limits their performance. We propose a self-alignment approach based on statistic moment matching to realize black-box domain adaptation. We construct a model for target domain in the initial stage of our work. Then, we put target data into source model API to obtain the pseudo-labels and divide the target data into high-confidence and low-confidence parts according to their pseudo-labels confidence. By matching the data distributions between these two parts and self-supervised learning on high-confidence part, the performance on both parts samples can be boosted respectively. Information maximization is also applied to the target data to further improve their classification performance. Experiment results confirm that our method achieves state-of-the-art performance.

Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

The fundamental challenge for the development of the wind energy conversion system is the natural instability of this resource, which is now at the forefront of the debate. The Internet of Things (IoT) has emerged as a smart and practical approach to renewable energy system management. In addition, the Internet of Things is an important study area with several possibilities for advancement and demanding work. To improve the dependability of wind energy conversion systems, researchers may simulate all potential wind turbine operating scenarios and design power electronics topologies and control algorithms appropriately. The proposed wind emulation platform is wind-turbine-oriented, with significantly increased traceability and efficiency when compared to existing systems. The wind emulation system is driven with real-time wind velocity data. The real-time and forecasted wind data is fetched from global nodes using an IoT cloud application programming interface. The wind emulation system is modelled in the VEE Pro platform and integrated with IoT cloud API and FPGA controller. Data obtained from IoT is converted into actionable information to analyse and investigate wind turbine performance. Also, effective security measures have been taken for ESP8266 communication with the Host Computer. Adapting the proposed IoT algorithm with an associated hardware prototype, the condition of the wind energy conversion system and the power generation capability shall be analyzed in a real-time environment.

New approach for cardiac patients based on pacemaker device

This current research work consists of the development of a new e-health application for the patients that have a cardiac problem and use a pacemaker. It allows: (1) heartbeat measure, (2) blood pressure surveillance, and (3) oxygen consumption surveillance. The main goal of our application is to assist the patients that have a pacemaker at any time and at any place. An IoT application based on remote sensing architecture is developed to avoid any critical state for the patients with pacemaker when catastrophe happened at any time and to react immediately. The different implemented sensor will transfer data from the patient’s body to the cloud API. Then, the different patients will be under permanent control and efficient monitor in smart environment. This solution is proposed for the well-being of each cardiac patient since that the doctor can know from the gathered data if everything is under control or not. If the heart does not beat regularly or the patient has trouble breathing, the responsible doctor and the patient in question will be alerted automatically. The Iot application is implemented and a real-simulations are done to highlight the importance of our e-health remote sensing-based application.

Integration of Data Mining as a Service on Cloud API via Cloud mining Algorithm

Integration of Data Mining as a Service on Cloud API via Cloud mining Algorithm

Weather Station IoT Educational Model Using Cloud Services

IoT technology is gaining more and more popularity in practice, as it collects, processes, evaluates and stores important measured data. The IoT is used every day in the work, in the home or smart houses or in public areas. It realizes the connectivity between real world and digital world which means, that it converts physical quantities of the real world in the form of analog signals into digital numbers stored in clauds. It is essential that students gain practical experience in the design and implementation of the IoT systems during their studies. The article first describes IoT issues and communication protocols used in IoT generally are closer described. Then the design and implementation of an educational model of IoT system - Weather station with the ThingSpeak cloud support is described. The created IoT model interconnects microcontroller programming, sensors and measuring, cloud API interfaces, MATLAB scripts which are useful to analyses the stored data, Windows and Android application developing.