Search

The growing demand for sustainable development brings a series of information technologies to help agriculture production. Especially, the emergence of machine learning applications, a branch of artificial intelligence, has shown multiple breakthroughs which can enhance and revolutionize plant pathology approaches. In recent years, machine learning has been adopted for leaf disease classification in both academic research and industrial applications. Therefore, it is enormously beneficial for researchers, engineers, managers, and entrepreneurs to have a comprehensive view about the recent development of machine learning technologies and applications for leaf disease detection. This study will provide a survey in different aspects of the topic including data, techniques, and applications. The paper will start with publicly available datasets. After that, we summarize common machine learning techniques, including traditional (shallow) learning, deep learning, and augmented learning. Finally, we discuss related applications. This paper would provide useful resources for future study and application of machine learning for smart agriculture in general and leaf disease classification in particular. [ABSTRACT FROM AUTHOR]

Humans use both auditory and facial cues to perceive speech, especially when auditory input is degraded, indicating a direct association between visual articulatory and acoustic speech information. This study investigates how well an audio signal of a word can be synthesized based on visual speech cues. Specifically, we synthesized audio waveforms of the vowels in monosyllabic English words from motion trajectories extracted from image sequences in the video recordings of the same words. The articulatory movements were recorded in two different speech styles: plain and clear. We designed a deep network trained on mouth landmark motion trajectories on a spectrogram and formant-based custom loss for different speech styles separately. Human and automatic evaluation show that our framework using visual cues can generate identifiable audio of the target vowels from distinct mouth landmark movements. Our results further demonstrate that intelligible audio can be synthesized from novel unseen talkers that were independent of the training data. [ABSTRACT FROM AUTHOR]

Fog computing has become a prominent paradigm in providing shared resources to serve different applications near the edge. Similar to other computing paradigms such as cloud and grid, in fog computing, service-level agreements (SLAs) are essential between fog providers and end-users to guarantee the quality of service (QoS). However, due to the unique characteristics of fog resources, such as being highly distributed and heterogeneous, with their dynamic nature having nonrestrictive provider participation, SLA management techniques and frameworks, which are available for Clouds and Grids, are not directly applicable. The availability of the resources in the cloud is much more controllable and predictable compared to fog. Moreover, due to the multiple ownership of fog infrastructure and unrestricted environment, autonomous end-devices are allowed to participate with different SLAs to serve the applications near the edge as a result is a lack of trust exists between the entities and managing and enforcing SLAs according to the application QoS in this environment is a complex task. Thus, the SLA management must be undertaken in a more trustworthy manner to ensure that agreement. To fill this gap, this paper proposes an automated SLA management framework for fog computing that utilizes Smart contracts and blockchain technology to monitor and enforce SLAs in a more trustworthy manner. The results obtained from the experiments, which were conducted in the blockchain private network, show that the framework can ensure precise and efficient SLAs enforcement in the fog. The performance of the proposed framework is better than existing work in terms of transaction cost and time. [ABSTRACT FROM AUTHOR]

Lipomas of the oral cavity are uncommon. Here we report a case of 85 year old female presenting with a progressively increasing large growth in the oropharynx which was diagnosed as lipoma on histopathology. The clinicoradiological and histopathological findings are discussed. To the best of our knowledge; this is one of the largest intraoral lipoma reported in India till date. The present case highlights the need to be aware of intraoral lipomas which can present as large growths at this unusual site so as to avoid any unwarranted aggressive surgery. [ABSTRACT FROM AUTHOR]

With the proliferation of intelligent devices in the Internet of Things (IoT), the security of IoT devices is becoming a major concern as an attacker can exploit network services via network capture or deployment of malicious devices. Therefore, IoT device authentication is a key security requirement. Conventional authentication approaches have proved insufficient in dealing with a large number of IoT devices because of various shortcomings, such as centrally designed architectures and expensive cryptographic primitives. Blockchain technology has led to the creation of decentralised IoT authentication solutions with their desirable characteristics. Due to the integration of IoT and Blockchain technology, however, performance is one of the significant challenges. Existing Blockchain-based authentication schemes create extra storage, computing and communication overheads on the Blockchain which restricts its use for different IoT networks. In this paper, we design an efficient decentralised Blockchain-based authentication scheme aimed at achieving a secure authentication for IoT devices using the token mechanism. We design the Ethereum Blockchain network model to implement our scheme, in which smart contracts provide secure connectivity between miner nodes and IoT devices. We demonstrate our proposed scheme with an IoT-based smart home case study, which shows significant performance in computation and communication overheads compared with existing Blockchain-based authentication schemes. The modelling, analysis and verification of the proposed scheme, using high level Petri nets (HLPNs) and Z3 SMT solver are also provided to prove correctness and to verify the claims of our model. [ABSTRACT FROM AUTHOR]

The interior noise in train carriages has become an important design parameter for passenger comfort with the proliferation of high-speed rail systems around the world, because passengers spend a few hours on these trains for traveling between cities spanning more than 1000 km apart. However, there are limited reported studies on comparing the carriage noise for different high-speed rail systems. In this study, the carriage noise of three high-speed (top speed between 200 and 300 km/h) and four express rail systems (top speed between 100 and 200 km/h) was recorded with an Android smartphone running Noise Explorer (an app we have developed for measuring noise) and analyzed on PC to benchmark differences among high-speed and express trains. The advantage of using a smartphone for measurements is to leverage citizen scientists for large-scale-simultaneous measurements. The results show that the average carriage noise level is between 64 and 75 dBA which is well below the typical legally acceptable limit of 85 dBA over an 8-h period for noise exposure. The average carriage noise level in dBC was found to be much higher than the corresponding average carriage noise level in dBA, indicating a significant presence of low-frequency content in the carriage noise. Finally, the carriage noise for a train in a tunnel was found to be higher than the same train on an open surface track by 4 to 7 dB. [ABSTRACT FROM AUTHOR]

In Web search, entity-seeking queries often trigger a special question answering (QA) system. It may use a parser to interpret the question to a structured query, execute that on a knowledge graph (KG), and return direct entity responses. QA systems based on precise parsing tend to be brittle: minor syntax variations may dramatically change the response. Moreover, KG coverage is patchy. At the other extreme, a large corpus may provide broader coverage, but in an unstructured, unreliable form. We present AQQUCN, a QA system that gracefully combines KG and corpus evidence. AQQUCN accepts a broad spectrum of query syntax, between well-formed questions to short "telegraphic" keyword sequences. In the face of inherent query ambiguities, AQQUCN aggregates signals from KGs and large corpora to directly rank KG entities, rather than commit to one semantic interpretation of the query. AQQUCN models the ideal interpretation as an unobservable or latent variable. Interpretations and candidate entity responses are scored as pairs, by combining signals from multiple convolutional networks that operate collectively on the query, KG and corpus. On four public query workloads, amounting to over 8000 queries with diverse query syntax, we see 5–16% absolute improvement in mean average precision (MAP), compared to the entity ranking performance of recent systems. Our system is also competitive at entity set retrieval, almost doubling F1 scores for challenging short queries. [ABSTRACT FROM AUTHOR]

In this paper, we present soundwalk of Miyajima and Itsukushima Shrine which was captured with a smartphone's built-in microphone calibrated against a typical type 1 sound level meter using NoiseExplorer our in-house developed Android app. Itsukushima Shrine (Itsukushima-jinja) is a Shinto shrine on the island of Itsukushima (popularly known as Miyajima), best known for its "floating" or "half-submerged" torii gate. It is in the southern part of Hatsukaichi in Hiroshima Prefecture in Japan. The shrine complex was listed as a UNESCO World Heritage Site in 1996. NoiseExplorer app records the audio from microphone as WAV files and the current location of the smartphone to log files. The WAV files were used to analyze sound pressure level and generate frequency spectrums and spectrograms, whereas log files were used to generate noise maps. These two capabilities of smartphones make them cheap and versatile means of recording and describing soundwalk or soundscape of any site in the world. We discuss the unique soundscape of the cross-channel ferry, the shrine itself, the footpath going up to the nearby mountain and the soundwalk captured using a smartphone. The overall noise level of 64.1 dBA is well below the typical noise level of other Japanese tourist attractions. [ABSTRACT FROM AUTHOR]

This paper presents a particle swarm optimization (PSO) technique to train an artificial neural network (ANN) for prediction of flank wear in drilling, and compares the network performance with that of the back propagation neural network (BPNN). This analysis is carried out following a series of experiments employing high speed steel (HSS) drills for drilling on mild steel workpieces, under different sets of cutting conditions and noting the root mean square (RMS) value of spindle motor current as well as the average flank wear in each case. The results show that the PSO trained ANN not only gives better prediction results and reduced computational times compared to the BPNN, it is also a more robust model, being free of getting trapped in local optimum solutions unlike the latter. Besides, it offers the advantages of a straight-forward logic, simple realization and underlying intelligence. [ABSTRACT FROM AUTHOR]

The recent development in Cloud computing has enabled the realization of delivering computing as an utility. Many industries such as Amazon and Google have started offering Cloud services on a 'pay as you go' basis. These advances have led to the evolution of the market infrastructure in the form of a Market Exchange (ME) that facilitates the trading between consumers and Cloud providers. Such market environment eases the trading process by aggregating IT services from a variety of sources, and allows consumers to easily select them. In this paper, we propose a light weight and platform independent ME framework called 'Mandi', which allows consumers and providers to trade computing resources according to their requirements. The novelty of Mandi is that it not only gives its users the flexibility in terms of negotiation protocol, but also allows the simultaneous coexistence of multiple trading negotiations. In this paper, we first present the requirements that motivated our design and discuss how these facilitate the trading of compute resources using multiple market models (also called negotiation protocols). Finally, we evaluate the performance of the first prototype of 'Mandi' in terms of its scalability. [ABSTRACT FROM AUTHOR]

We report an investigation of epitaxial germanium grown by chemical vapor deposition (CVD) on arsenic-terminated (211)Si, which is the preferred substrate in the USA for fabrication of night-vision devices based on mercury cadmium telluride (MCT) grown by molecular-beam epitaxy (MBE). The films were characterized by scanning electron microscopy (SEM), atomic force microscopy (AFM), cross-sectional transmission electron microscopy (XTEM), and x-ray diffraction (XRD). Arsenic passivation was found to be effective in preventing cross-contamination of unwanted residual species present inside the reactor chamber and also in prolonging the evolution of layer-by-layer growth of Ge for significantly more monolayers than on nonpassivated Si. The two-dimensional (2D) to three-dimensional (3D) transition resulted in Ge islands, the density and morphology of which showed a clear distinction between passivated and nonpassivated (211)Si. Finally, thick Ge layers (∼250 nm) were grown at 525°C and 675°C with and without As passivation, where the layers grown with As passivation resulted in higher crystal quality and smooth surface morphology. [ABSTRACT FROM AUTHOR]

Rett's syndrome (RTT) is an X-chromosome-linked autism spectrum disorder caused by loss of function of the transcription factor methyl-CpG-binding protein 2 (MeCP2). Although MeCP2 is expressed in most tissues, loss of MeCP2 expression results primarily in neurological symptoms. Earlier studies suggested the idea that RTT is due exclusively to loss of MeCP2 function in neurons. Although defective neurons clearly underlie the aberrant behaviours, we and others showed recently that the loss of MECP2 from glia negatively influences neurons in a non-cell-autonomous fashion. Here we show that in globally MeCP2-deficient mice, re-expression of Mecp2 preferentially in astrocytes significantly improved locomotion and anxiety levels, restored respiratory abnormalities to a normal pattern, and greatly prolonged lifespan compared to globally null mice. Furthermore, restoration of MeCP2 in the mutant astrocytes exerted a non-cell-autonomous positive effect on mutant neurons in vivo, restoring normal dendritic morphology and increasing levels of the excitatory glutamate transporter VGLUT1. Our study shows that glia, like neurons, are integral components of the neuropathology of RTT, and supports the targeting of glia as a strategy for improving the associated symptoms. [ABSTRACT FROM AUTHOR]

This study compares the performance of backpropagation neural network (BPNN) and radial basis function network (RBFN) in predicting the flank wear of high speed steel drill bits for drilling holes on mild steel and copper work pieces. The validation of the methodology is carried out following a series of experiments performed over a wide range of cutting conditions in which the effect of various process parameters, such as drill diameter, feed-rate, spindle speed, etc. on drill wear has been considered. Subsequently, the data, divided suitably into training and testing samples, have been used to effectively train both the backpropagation and radial basis function neural networks, and the individual performance of the two networks is then analyzed. It is observed that the performance of the RBFN fails to match that of the BPNN when the network complexity and the amount of data available are the constraining factors. However, when a simpler training procedure and reduced computational times are required, then RBFN is the preferred choice. [ABSTRACT FROM AUTHOR]

This study contributes to the overarching Indian Environmental Radiation Monitoring Network initiative, concentrating on the continual assessment of the gamma absorbed dose rate in outdoor air attributable to natural gamma radiation in Kolkata, West Bengal, India. Over a span of fourteen years (2011–2024), data were gathered from 32 monitoring sites within the city utilizing permanently installed Geiger-Mueller detector-based environmental radiation monitors. This paper offers an analysis of extensive findings from long-term monitoring efforts. The absorbed gamma dose rate in outdoor air varied among the monitoring sites, ranging from 78 ± 5 to 137 ± 4 nGy h− 1, with a mean value of 107 ± 12 nGy h− 1. The estimated mean annual effective dose due to outdoor natural gamma radiation ranged from 0.10 ± 0.01 to 0.17 ± 0.01 mSv y− 1, with an overall mean of 0.13 ± 0.01 mSv y−1. [ABSTRACT FROM AUTHOR]

Matrix operations are fundamental to a wide range of scientific applications such as Graph Theory, Linear Equation System, Image Processing, Geometric Optics, and Probability Analysis. As the workload in these applications has increased, the sizes of matrices involved have also significantly increased. Parallel execution of matrix operations in existing cluster-based systems performs effectively for relatively small matrices but significantly suffers as matrices become larger due to limited resources. Cloud Computing offers scalable resources to handle this limitation; however, the benefits of having access to almost-infinite scalable resources in the Cloud also come with challenges of ensuring time and resource-efficient matrix operations. To the best of our knowledge, there is no specific Cloud service that optimizes the efficiency of matrix operations on Cloud infrastructure. To address this gap and offer convenient service of matrix operations, the paper proposes a novel scalable service framework called Scalable Matrix Operation as a Service. Our framework uses Dynamic Matrix Partition techniques, based on matrix operation and sizes, to achieve efficient work distribution, and scales based on demand to achieve time and resource-efficient operations. The framework also embraces the basic features of security, fault tolerance, and reliability. Experimental results show that the adopted dynamic partitioning technique ensures faster and better performance when compared to the existing static partitioning technique. [ABSTRACT FROM AUTHOR]