Search

The Indian economy is greatly influenced by the Banana Industry, necessitating advancements in agricultural farming. Recent research emphasizes the imperative nature of addressing diseases that impact Banana Plants, with a particular focus on early detection to safeguard production. The urgency of early identification is underscored by the fact that diseases predominantly affect banana plant leaves. Automated systems that integrate machine learning and deep learning algorithms have proven to be effective in predicting diseases. This manuscript examines the prediction and detection of diseases in banana leaves, exploring various diseases, machine learning algorithms, and methodologies. The study makes a contribution by proposing two approaches for improved performance and suggesting future research directions. In summary, the objective is to advance understanding and stimulate progress in the prediction and detection of diseases in banana leaves. The need for enhanced disease identification processes is highlighted by the results of the survey. Existing models face a challenge due to their lack of rotation and scale invariance. While algorithms such as random forest and decision trees are less affected, initially convolutional neural networks (CNNs) is considered for disease prediction. Though the Convolutional Neural Network models demonstrated impressive accuracy in many research but it lacks in invariance to scale and rotation. Moreover, it is observed that due its inherent design it cannot be combined with feature extraction methods to identify the banana leaf diseases. Due to this reason two alternative models that combine ANN with scale-invariant Feature transform (SIFT) model or histogram of oriented gradients (HOG) combined with local binary patterns (LBP) model are suggested. The first model ANN with SIFT identify the disease by using the activation functions to process the features extracted by the SIFT by distinguishing the complex patterns. The second integrate the combined features of HOG and LBP to identify the disease thus by representing the local pattern and gradients in an image. This paves a way for the ANN to learn and identify the banana leaf disease. Moving forward, exploring datasets in video formats for disease detection in banana leaves through tailored machine learning algorithms presents a promising avenue for research., (© 2024. The Author(s).)

The purpose of the study is to develop an approach to planning the trajectory of the manipulator robot using an intelligent system based on neural networks. For this purpose, the work considered the processes of planning and deploying the movement of the robot. The analysis of existing methods of planning the movement of manipulator robots and the review of intelligent control systems made it possible to obtain a complete picture of the current state of this issue. A system is proposed that can perceive the environment and control the movement of the robot by generating the correct control commands. For this, 3 tasks were solved, namely: analysis of the environment in order to determine its features, determination of the trajectory in order to neutralize the collision and determination of controlled influences for the executive authorities in order to implement the movement. The functionality and structure of the neural network for solving each of the tasks are proposed. [ABSTRACT FROM AUTHOR]

In the era of automated and digitalized information, advanced computer applications deal with a major part of the data that comprises audio-related information. Advancements in technology have ushered in a new era where cutting-edge devices can deliver comprehensive insights into audio content, leveraging sophisticated algorithms such such as Mel Frequency Cepstral Coefficients (MFCCs) and Short-Time Fourier Transform (STFT) to extract and provide pertinent information. Our study helps in not only efficient audio file management and audio file retrievals but also plays a vital role in security, the robotics industry, and investigations. Beyond its industrial applications, our model exhibits remarkable versatility in the corporate sector, particularly in tasks like siren sound detection and more. Embracing this capability holds the promise of catalyzing the development of advanced automated systems, paving the way for increased efficiency and safety across various corporate domains. The primary aim of our experiment is to focus on creating highly efficient audio classification models that can be seamlessly automated and deployed within the industrial sector, addressing critical needs for enhanced productivity and performance. Despite the dynamic nature of environmental sounds and the presence of noises, our presented audio classification model comes out to be efficient and accurate. The novelty of our research work reclines to compare two different audio datasets having similar characteristics and revolves around classifying the audio signals into several categories using various machine learning techniques and extracting MFCCs and STFTs features from the audio signals. We have also tested the results after and before the noise removal for analyzing the effect of the noise on the results including the precision, recall, specificity, and F1-score. Our experiment shows that the ANN model outperforms the other six audio models with the accuracy of 91.41% and 91.27% on respective datasets. [ABSTRACT FROM AUTHOR]

Objective. Development of automated network attack detection systems capable of adapting to the ever-changing nature of network attacks and new types of threats. Such systems should be based on machine learning algorithms and models that are able to identify complex dependencies between data in the learning process.Method. To train the models, a sample with signs of normal and abnormal traffic was prepared, and it was thinned and balanced as a result of preliminary statistical analysis. Five machine learning algorithms were selected and tested, both on a training set of features and on a real test set obtained experimentally. Based on the results of the experiments, a random forest classifier was selected, which showed the best results.Result. A model for detecting network intrusions has been developed, which showed a detection accuracy of 0.99 on real traffic.Conclusion. It is shown that a machine learning-based network intrusion detection system can solve the problem of flexible protection that could adapt to the ever-changing nature of network attacks, since one of the most important advantages of machine learning in detecting network intrusions is the ability to learn the signs of attacks and identify cases that are uncharacteristic of those that were observed earlier.

In the dynamic world of financial markets, the prediction of stock performance and bitcoin trading is undergoing a significant transformation due to the integration of advanced technologies and novel methodologies. The incorporation of Transformer models alongside Time Embeddings significantly improves the precision of stock market predictions by effectively capturing intricate temporal relationships and mitigating the presence of overly simplistic assumptions. The integration of real-time social media data with sentiment analysis based on BERT provides significant value in understanding investor sentiment. Additionally, the application of language model pre-training, as exemplified by BERT, brings about a transformative impact on text classification for predicting stock prices. Within the domain of cryptocurrency, sophisticated algorithms such as Transformers, Long Short-Term Memory (LSTM), Deep Convolutional LSTM (DC-LSTM), and Neural Networks (NN) have demonstrated enhanced capabilities in predicting price movements. These algorithms are further bolstered by the implementation of a comprehensive trading strategy. Automated systems for bitcoin trading introduce elements of personalization and adaptability to the trading process, thereby facilitating broader access to a diverse group of traders. The progress highlights the significant importance of the integration of technology and methodologies in the field of financial analysis. This integration enables investors and traders to possess the necessary resources for making well-informed choices within the ever-changing landscape of financial markets. [ABSTRACT FROM AUTHOR]

The studies related to the coherent structures in the atmosphere, using Doppler wind lidar observations, so far have relied on the manual detection and classification of the structures in the lidar images, making this process time- consuming. We developed an automated classification that is based on texture analysis parameters and the quadratic discriminant analysis algorithm for the detection of medium-to-large fluctuations and coherent structures recorded by single Doppler wind lidar quasi-horizontal scans. The algorithm classified a training dataset of 150 cases into four types of patterns, namely, streaks (narrow stripes), rolls (wide stripes), thermals (enclosed areas), and ''others'' (impossible to classify), with 91% accuracy. Subsequently, we applied the trained algorithm to a dataset of 4577 lidar scans recorded in Paris, atop a 75-m tower for a 2-month period (September--October 2014). The current study assesses the quality of the classification by examining the physical properties of the classified cases. The results show a realistic classification of the data: with rolls and thermals cases mostly classified concurrently with a well-developed atmospheric boundary layer and the streaks cases associated with nocturnal low-level jets events. Furthermore, rolls and streaks cases were mostly observed under moderate or high wind conditions. The detailed analysis of a 4-day period reveals the transition between the types. The analysis of the space spectra in the direction transverse to the mean wind, during these four days, revealed streak spacing of 200-400m and roll sizes, as observed in the lower level of the mixed layer, of approximately 1 km. [ABSTRACT FROM AUTHOR]

Nanomaterials are at the core of many scientific discoveries in catalysis, energy and healthcare to name a few. However, their deployment is limited by the lack of reproducible and precise manufacturing technologies on‐demand. In this work, a precision automated technology is demonstrated for nanoparticles synthesis with wide‐range tunable sizes (≈4–100 nm). Dial‐a‐particle capabilities are achieved by a combination of a fast integrated multipoint particle sizing combined with a "plug‐n‐play" modular platform with reactors in series, distributed feed and in situ multipoint analysis. Real‐time early growth information accurately predicts the resulting particle properties. Such real‐time simple feedback control can overcome repeatability and stability issues associated with controllable (e.g., conditions) and uncontrollable (e.g., fouling, ageing, and impurities) variations leading to self‐regulated, highly stable multistage systems with no human intervention even with long residence times (from a few minutes to hours). This is a paradigm shift from machine learning (ML) methodologies, which are restricted to trained networks with rich data sets, impractical in non‐reproducible processes and limited to short residence times (e.g., within few minutes). The approach is demonstrated for plasmonic silver and gold nanoparticles showing agile control within minutes, opening the door for automation of more complex multistage procedures such as composites, multielement materials, and particle functionalization. [ABSTRACT FROM AUTHOR]

Blizzard conditions occur regularly in the Canadian Arctic, with high impact on travel and life there. These extreme conditions are challenging to forecast for this vast domain because the observation network is sparse and remote sensing coverage is limited. To establish occurrence statistics we analyzed aviation routine weather reports (METARs) from Canadian Arctic stations between October and May 2014–18. Blizzard conditions occur most frequently in open tundra east and north of the boreal forest boundary, with the highest frequency found on the northwest side of Hudson Bay and over flat terrain in central Baffin Island. Except in sheltered locations, the reported cause of reduced visibility is blowing snow without precipitating snow in about one-half to two-thirds of METARs made by a human observer, even higher at some stations. We produce three products that forecast blizzard conditions from postprocessed NWP model output. The blizzard potential (BP), generated from expert’s rules, is intended for warning well in advance of areas where blizzard conditions may develop. A second product (BH) stems from regression equations for the probability of visibility ≤ 1 km in blowing snow and/or concurrent snow derived by Baggaley and Hanesiak. A third product (RF), generated with the random forest ensemble classification algorithm, makes a consensus YES/NO forecast for blizzard conditions. We describe the products, provide verification, and show forecasts for a significant blizzard event. Receiver operator characteristic curves and critical success index scores show RF forecasts have greater accuracy than BP and BH forecasts at all lead times. [ABSTRACT FROM AUTHOR]

Featured Application: This study proposes a method that enables the automatic application of image cross-correlation when monitoring any displacements in the geosciences. As such, it solves one of the main current image processing issues: The requirement of manual image selection. The method reduces the need for extensive financial and human resources when conducting surveys, and it can be applied in a preventive warning context. In Earth Science, image cross-correlation (ICC) can be used to identify the evolution of active processes. However, this technology can be ineffective, because it is sometimes difficult to visualize certain phenomena, and surface roughness can cause shadows. In such instances, manual image selection is required to select images that are suitably illuminated, and in which visibility is adequate. This impedes the development of an autonomous system applied to ICC in monitoring applications. In this paper, the uncertainty introduced by the presence of shadows is quantitatively analysed, and a method suitable for ICC applications is proposed: The method automatically selects images, and is based on a supervised classification of images using the support vector machine. According to visual and illumination conditions, the images are divided into three classes: (i) No visibility, (ii) direct illumination and (iii) diffuse illumination. Images belonging to the diffuse illumination class are used in cross-correlation processing. Finally, an operative procedure is presented for applying the automated ICC processing chain in geoscience monitoring applications. [ABSTRACT FROM AUTHOR]

In Earth Science, image cross-correlation (ICC) can be used to identify the evolution of active processes. However, this technology can be ineffective, because it is sometimes difficult to visualize certain phenomena, and surface roughness can cause shadows. In such instances, manual image selection is required to select images that are suitably illuminated, and in which visibility is adequate. This impedes the development of an autonomous system applied to ICC in monitoring applications. In this paper, the uncertainty introduced by the presence of shadows is quantitatively analysed, and a method suitable for ICC applications is proposed: The method automatically selects images, and is based on a supervised classification of images using the support vector machine. According to visual and illumination conditions, the images are divided into three classes: (i) No visibility, (ii) direct illumination and (iii) diffuse illumination. Images belonging to the diffuse illumination class are used in cross-correlation processing. Finally, an operative procedure is presented for applying the automated ICC processing chain in geoscience monitoring applications.

The recent National Students Survey showed that feedback to students was an ongoing problem in Higher Education. This paper reports on the extension of our past research into the provision of automated feedback for objective testing. In the research presented here, the system has been further developed for marking practical and essay questions and providing automated feedback. Recent research at the University of Hertfordshire was able to show that learners and tutors accept and value our automated feedback approach based on objective tests and Computer Adaptive Testing. The research reported in this paper is an important extension to this work. The automated feedback system developed for objective testing has been extended to include practical testing and essay type questions. The automated feedback system, which can be used within any subject area, is based on a simple marking scheme created by the subject tutor as a text file according to a simple template. Marks for each option and a set of feedback statements are held within a database on a computer. As marks are awarded for each question by the teacher an individual feedback file is created automatically for each learner. Teachers may also add and modify comments to each learner and save additional feedback to the database for later use. Each individual feedback file was emailed automatically to learners. The development of the system is explained in the paper and testing and evaluation with 350 first year (1 final practical test), 120 second year (1 written and 1 practical tests) and 100 final year (1 final practical test) undergraduate Computer Science students is reported. It was found that the time to mark practical and essay type tests was reduced by more than 30% in all cases compared to previous years. More importantly it was possible to provide good quality individual feedback to learners rapidly. Feedback was delivered to all within three weeks of the test submission date. In end of module tests it was very beneficial indeed as it had proven difficult to provide feedback in the past after modules had ended. Examples of the feedback provided are presented in the paper and the development of the system using a user-centred approach based on student and staff evaluation is explained. The comments of staff teaching on these modules and a sample of students who took part in this series of evaluations of the system are presented. The results of these evaluations were very positive and are reported in the paper, showing the changes that were made to the system at each iteration of the development cycle. The provision of fast effective feedback is vital and this system was found to be an important addition to the tools available. [ABSTRACT FROM AUTHOR]

Summary: Objective: AutoNRT™ is an automated system that measures electrically evoked compound action potential (ECAP) thresholds from the auditory nerve with the Nucleus® Freedom™ cochlear implant. ECAP thresholds along the electrode array are useful in objectively fitting cochlear implant systems for individual use. This paper provides the first detailed description of the AutoNRT algorithm and its expert systems, and reports the clinical success of AutoNRT to date. Methods: AutoNRT determines thresholds by visual detection, using two decision tree expert systems that automatically recognise ECAPs. The expert systems are guided by a dataset of 5393 neural response measurements. The algorithm approaches threshold from lower stimulus levels, ensuring recipient safety during postoperative measurements. Intraoperative measurements use the same algorithm but proceed faster by beginning at stimulus levels much closer to threshold. When searching for ECAPs, AutoNRT uses a highly specific expert system (specificity of 99% during training, 96% during testing; sensitivity of 91% during training, 89% during testing). Once ECAPs are established, AutoNRT uses an unbiased expert system to determine an accurate threshold. Throughout the execution of the algorithm, recording parameters (such as implant amplifier gain) are automatically optimised when needed. Results: In a study that included 29 intraoperative and 29 postoperative subjects (a total of 418 electrodes), AutoNRT determined a threshold in 93% of cases where a human expert also determined a threshold. When compared to the median threshold of multiple human observers on 77 randomly selected electrodes, AutoNRT performed as accurately as the ‘average’ clinician. Conclusions: AutoNRT has demonstrated a high success rate and a level of performance that is comparable with human experts. It has been used in many clinics worldwide throughout the clinical trial and commercial launch of Nucleus Custom Sound™ Suite, significantly streamlining the clinical procedures associated with cochlear implant use. [Copyright &y& Elsevier]

Automated systems allow forensic practitioners to perform analysis tasks that would otherwise be infeasible. However, unless the outputs of such systems are critically evaluated using a scientifically-based framework, there is a risk of undetected errors or bias resulting in wrong decisions. Furthermore, decisions based on automated system outputs that are not well understood or clearly explainable could violate fundamental human rights. These risks can apply to any automated system that supports forensic analysis, and are raised when machine learning is involved. This work presents a framework based on principles of scientific interpretation, and provides an evaluation hierarchy for automated systems, including machine learning approaches, to strengthen forensic conclusions. Specifically, three levels of evaluation are presented: performance, understandability and forensic evaluation. Approaches to clearly conveying the weight of forensic evaluations are discussed. Each level of evaluation is demonstrated in relation to actual automated systems. Finally, requirements for designing automated systems supporting forensic analysis are proposed, and future work is discussed. • Design requirements for automated systems supporting forensic analysis using a framework of scientific reasoning. • Hierarchy for evaluating outputs from automated systems, including AI approaches, to strengthen forensic conclusions. • Practical examples of how three levels of evaluation can be applied to the outputs of automated systems. [ABSTRACT FROM AUTHOR]