A Comparison of YOLO Family for Apple Detection and Counting in Orchards

In agricultural production and breeding, implementing automatic picking robot in orchard farming to reduce human labour and error is challenging. The core function of it is automatic identification based on machine vision. This paper focuses on apple detection and counting in orchards and implements several deep learning methods. Extensive datasets are used and a semi-automatic annotation method is proposed. The proposed deep learning models are in state-of-the-art YOLO family. In view of the essence of the models with various backbones, a multi-dimensional comparison in details is made in terms of counting accuracy, mAP and model memory, laying the foundation for realising automatic precision agriculture.

Improving Subjective Bias Detection Using Bidirectional Encoder Representations from Transformers and Bidirectional Long Short-Term Memory

Detecting subjectively biased statements is a vital task. This is because this kind of bias, when present in the text or other forms of information dissemination media such as news, social media, scientific texts, and encyclopedias, can weaken trust in the information and stir conflicts amongst consumers. Subjective bias detection is also critical for many Natural Language Processing (NLP) tasks like sentiment analysis, opinion identification, and bias neutralization. Having a system that can adequately detect subjectivity in text will boost research in the above-mentioned areas significantly. It can also come in handy for platforms like Wikipedia, where the use of neutral language is of importance. The goal of this work is to identify the subjectively biased language in text on a sentence level. With machine learning, we can solve complex AI problems, making it a good fit for the problem of subjective bias detection. A key step in this approach is to train a classifier based on BERT (Bidirectional Encoder Representations from Transformers) as upstream model. BERT by itself can be used as a classifier; however, in this study, we use BERT as data preprocessor as well as an embedding generator for a Bi-LSTM (Bidirectional Long Short-Term Memory) network incorporated with attention mechanism. This approach produces a deeper and better classifier. We evaluate the effectiveness of our model using the Wiki Neutrality Corpus (WNC), which was compiled from Wikipedia edits that removed various biased instances from sentences as a benchmark dataset, with which we also compare our model to existing approaches. Experimental analysis indicates an improved performance, as our model achieved state-of-the-art accuracy in detecting subjective bias. This study focuses on the English language, but the model can be fine-tuned to accommodate other languages.

Identification of Configuration Space Singularities with Local Real Algebraic Geometry

We address the question of identifying the configuration space singularities of linkages, i.e., points where the configuration space is not locally a submanifold of Euclidean space. Because the configuration space cannot be smoothly parameterized at such points, these singularity types have a significantly negative impact on the kinematics of the linkage. It is known that Jacobian methods do not provide sufficient conditions for the existence of CS-singularities. Herein, we present several additional algebraic criteria that provide the sufficient conditions. Further, we use those criteria to analyze certain classes of planar linkages. These examples will also show how the presented criteria can be checked using algorithmic methods.

A Structural Support Vector Machine Approach for Biometric Recognition

Face is a non-intrusive strong biometrics for identification of original and dummy facial by different artificial means. Face recognition is extremely important in the contexts of computer vision, psychology, surveillance, pattern recognition, neural network, content based video processing. The availability of a widespread face database is crucial to test the performance of these face recognition algorithms. The openly available face databases include face images with a wide range of poses, illumination, gestures and face occlusions but there is no dummy face database accessible in public domain. This paper presents a face detection algorithm based on the image segmentation in terms of distance from a fixed point and template matching methods. This proposed work is having the most appropriate number of nodal points resulting in most appropriate outcomes in terms of face recognition and detection. The time taken to identify and extract distinctive facial features is improved in the range of 90 to 110 sec. with the increment of efficiency by 3%.

Automated Method Time Measurement System for Redesigning Dynamic Facility Layout

The dynamic facility layout problem is a really critical issue in the competitive industrial market; thus, solving this problem requires robust design and effective simulation systems. The sustainable simulation requires inputting reliable and accurate data into the system. So this paper describes an automated system integrated into the real environment to measure the duration of the material handling operations, collect the data in real-time, and determine the variances between the actual and estimated time schedule of the operations in order to update the simulation software and redesign the facility layout periodically. The automated method- time measurement system collects the real data through using Radio Frequency-Identification (RFID) and Internet of Things (IoT) technologies. Hence, attaching RFID- antenna reader and RFID tags enables the system to identify the location of the objects and gathering the time data. The real duration gathered will be manipulated by calculating the moving average duration of the material handling operations, choosing the shortest material handling path, and then updating the simulation software to redesign the facility layout accommodating with the shortest/real operation schedule. The periodic simulation in real-time is more sustainable and reliable than the simulation system relying on an analysis of historical data. The case study of this methodology is in cooperation with a workshop team for producing mechanical parts. Although there are some technical limitations, this methodology is promising, and it can be significantly useful in the redesigning of the manufacturing layout.

Thresholding Approach for Automatic Detection of Pseudomonas aeruginosa Biofilms from Fluorescence in situ Hybridization Images

Pseudomonas aeruginosa is an opportunistic pathogen that forms surface-associated microbial communities (biofilms) on artificial implant devices and on human tissue. Biofilm infections are difficult to treat with antibiotics, in part, because the bacteria in biofilms are physiologically heterogeneous. One measure of biological heterogeneity in a population of cells is to quantify the cellular concentrations of ribosomes, which can be probed with fluorescently labeled nucleic acids. The fluorescent signal intensity following fluorescence in situ hybridization (FISH) analysis correlates to the cellular level of ribosomes. The goals here are to provide computationally and statistically robust approaches to automatically quantify cellular heterogeneity in biofilms from a large library of epifluorescent microscopy FISH images. In this work, the initial steps were developed toward these goals by developing an automated biofilm detection approach for use with FISH images. The approach allows rapid identification of biofilm regions from FISH images that are counterstained with fluorescent dyes. This methodology provides advances over other computational methods, allowing subtraction of spurious signals and non-biological fluorescent substrata. This method will be a robust and user-friendly approach which will enable users to semi-automatically detect biofilm boundaries and extract intensity values from fluorescent images for quantitative analysis of biofilm heterogeneity.

Improved Rare Species Identification Using Focal Loss Based Deep Learning Models

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

A Visual Analytics Tool for the Structural Health Monitoring of an Aircraft Panel

Aerospace, mechanical, and civil engineering infrastructures can take advantages from damage detection and identification strategies in terms of maintenance cost reduction and operational life improvements, as well for safety scopes. The challenge is to detect so called “barely visible impact damage” (BVID), due to low/medium energy impacts, that can progressively compromise the structure integrity. The occurrence of any local change in material properties, that can degrade the structure performance, is to be monitored using so called Structural Health Monitoring (SHM) systems, in charge of comparing the structure states before and after damage occurs. SHM seeks for any "anomalous" response collected by means of sensor networks and then analyzed using appropriate algorithms. Independently of the specific analysis approach adopted for structural damage detection and localization, textual reports, tables and graphs describing possible outlier coordinates and damage severity are usually provided as artifacts to be elaborated for information extraction about the current health conditions of the structure under investigation. Visual Analytics can support the processing of monitored measurements offering data navigation and exploration tools leveraging the native human capabilities of understanding images faster than texts and tables. Herein, a SHM system enrichment by integration of a Visual Analytics component is investigated. Analytical dashboards have been created by combining worksheets, so that a useful Visual Analytics tool is provided to structural analysts for exploring the structure health conditions examined by a Principal Component Analysis based algorithm.

Signal and Harmonic Analysis of a Compressor Blade for Identification of the Nonlinear Frequency Vibration

High-speed turbomachine can experience significant centrifugal and gas bending loads. As a result, the compressor blades must be able to resist high-frequency oscillations due to surge or stall condition in flow field dynamics. In this paper, vibration characteristics of the 6th stage blade compressor have been examined in detail with, using 3-D finite element (FE) methods. The primary aim of this article is to gain an understanding of nonlinear vibration induced in the blade against different loading conditions. The results indicate the nonlinear behavior of the blade as a result of the amplitude of resonances or material properties. Since one of the leading causes of turbine blade failure is high cycle fatigue, simulations were started by specifying the stress distribution in the blade due to the centrifugal rotation. Next, resonant frequencies and critical speeds of the blade were defined by modal analysis. Finally, the harmonic analysis was simulated on the blades.

Harmonic Pollution Control of the Electrical Network by Three-Phase Shunt Active Filter: Comparative Study of Controls, by Hysteresis and by Duty Cycle Modulation

This paper deals with the harmonic decontamination of current in an electrical grid by an active shunt filter in order to improve power quality. The contribution of this paper is mainly based on the proposal of a control strategy for an active filter based on Duty Cycle Modulation (DCM). First, three-monophase method is applied for the identification of disturbing currents. A Simulink model of this method is given for one phase of the grid. Secondly, two orders were designed: the first one is the Hysteresis Control and the second one is the DCM Control. Finally, a comparative study of the two controls was performed. The results obtained show a significant improvement in the rate of harmonic distortion for both controls. The harmonic distortion for the Hysteresis control is limited by the non-controllability of the switching frequencies of the inverter's switches and reduces the harmonic distortion rate (THD) to 3.12% as opposed to the DCM control which limits the THD to 2.82% which makes it better.

Privacy Protection Principles of Omnichannel Approach

The advent of the Internet, mobile devices and social media is revolutionizing the experience of retail customers by linking multiple sources through various channels. Omnichannel retailing is a retailing that combines multiple channels to allow customers to seamlessly leverage all the distribution information online and offline while shopping. Therefore, today data are an asset more critical than ever for all organizations. Nonetheless, because of its heterogeneity through platforms, developers are currently facing difficulties in dealing with personal data. Considering the possibilities of omnichannel communication, this paper presents channel categorization that could enhance the customer experience of omnichannel center called hyper center. The purpose of this paper is fundamentally to describe the connection between the omnichannel hyper center and the customer, with particular attention to privacy protection. The first phase was finding the most appropriate channels of communication for hyper center. Consequently, a selection of widely used communication channels has been identified and analyzed with regard to the effect requirements for optimizing user experience. The evaluation criteria are divided into 3 groups: general, user profile and channel options. For each criterion the weight of importance for omnichannel communication was defined. The most important thing was to consider how the hyper center can make user identification while respecting the privacy protection requirements. The study carried out also shows what customer experience across digital networks would look like, based on an omnichannel approach owing to privacy protection principles.

Development of Fake News Model Using Machine Learning through Natural Language Processing

Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Naïve Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.

Spatial-Temporal Awareness Approach for Extensive Re-Identification

Recent development of AI and edge computing plays a critical role to capture meaningful events such as detection of an unattended bag. One of the core problems is re-identification across multiple CCTVs. Immediately following the detection of a meaningful event is to track and trace the objects related to the event. In an extensive environment, the challenge becomes severe when the number of CCTVs increases substantially, imposing difficulties in achieving high accuracy while maintaining real-time performance. The algorithm that re-identifies cross-boundary objects for extensive tracking is referred to Extensive Re-Identification, which emphasizes the issues related to the complexity behind a great number of CCTVs. The Spatial-Temporal Awareness approach challenges the conventional thinking and concept of operations which is labor intensive and time consuming. The ability to perform Extensive Re-Identification through a multi-sensory network provides the next-level insights – creating value beyond traditional risk management.

Automatic Product Identification Based on Deep-Learning Theory in an Assembly Line

Automated object recognition and identification systems are widely used throughout the world, particularly in assembly lines, where they perform quality control and automatic part selection tasks. This article presents the design and implementation of an object recognition system in an assembly line. The proposed shapes-color recognition system is based on deep learning theory in a specially designed convolutional network architecture. The used methodology involve stages such as: image capturing, color filtering, location of object mass centers, horizontal and vertical object boundaries, and object clipping. Once the objects are cut out, they are sent to a convolutional neural network, which automatically identifies the type of figure. The identification system works in real-time. The implementation was done on a Raspberry Pi 3 system and on a Jetson-Nano device. The proposal is used in an assembly course of bachelor’s degree in industrial engineering. The results presented include studying the efficiency of the recognition and processing time.

Analyzing Microblogs: Exploring the Psychology of Political Leanings

Microblogging has become increasingly popular for commenting on current events, spreading gossip, and encouraging individualism--which favors its low-context communication channel. These social media (SM) platforms allow users to express opinions while interacting with a wide range of populations. Hashtags allow immediate identification of like-minded individuals worldwide on a vast array of topics. The output of the analytic tool, Linguistic Inquiry and Word Count (LIWC)--a program that associates psychological meaning with the frequency of use of specific words--may suggest the nature of individuals’ internal states and general sentiments. When applied to groupings of SM posts unified by a hashtag, such information can be helpful to community leaders during periods in which the forming of public opinion happens in parallel with the unfolding of political, economic, or social events. This is especially true when outcomes stand to impact the well-being of the group. Here, we applied the online tools, Google Translate and the University of Texas’s LIWC, to a 90-posting sample from a corpus of Colombian Spanish microblogs. On translated disjoint sets, identified by hashtag as being authored by advocates of voting “No,” advocates voting “Yes,” and entities refraining from hashtag use, we observed the value of LIWC’s Tone feature as distinguishing among the categories and the word “peace,” as carrying particular significance, due to its frequency of use in the data.

A Goal-Driven Crime Scripting Framework

Crime scripting is a simple and effective crime modeling technique that aims to improve understanding of security analysts about security and crime incidents. Low-quality scripts provide a wrong, incomplete, or sophisticated understanding of the crime commission process, which oppose the purpose of their application, e.g., identifying effective and cost-efficient situational crime prevention (SCP) measures. One important and overlooked factor in generating quality scripts is the crime scripting method. This study investigates the problems within the existing crime scripting practices and proposes a crime scripting approach that contributes to generating quality crime scripts. It was validated by experienced crime scripters. This framework helps analysts develop better crime scripts and contributes to their effective application, e.g., SCP measures identification or policy-making.

Influence of Environmental Temperature on Dairy Herd Performance and Behaviour

The objective of this study was to determine the effects of environmental stressors on the performance of lactating dairy cows and discuss some future trends. There exists a relationship between the meteorological data and milk yield prediction accuracy in pasture-based dairy systems. New precision technologies are available and are being developed to improve the sustainability of the dairy industry. Some of these technologies focus on welfare of individual animals on dairy farms. These technologies allow the automatic identification of animal behaviour and health events, greatly increasing overall herd health and yield while reducing animal health inspection demands and long-term animal healthcare costs. The data set consisted of records from 489 dairy cows at two dairy farms and temperature measured from the nearest meteorological weather station in 2018. The effects of temperature on milk production and behaviour of animals were analyzed. The statistical results indicate different effects of temperature on milk yield and behaviour. The “comfort zone” for animals is in the range 10 °C to 20 °C. Dairy cows out of this zone had to decrease or increase their metabolic heat production, and it affected their milk production and behaviour.

Exploring the Potential of Chatbots in Higher Education: A Preliminary Study

We report upon a study introducing a chatbot to develop learning communities at a London University, with a largely international student base. The focus of the chatbot was twofold; to ease the transition for students into their first year of university study, and to increase study engagement. Four learning communities were created using the chatbot; level 3 foundation, level 4 undergraduate, level 6 undergraduate and level 7 post-graduate. Students and programme leaders were provided with access to the chat bot via mobile app prior to their study induction and throughout the autumn term of 2019. At the end of the term, data were collected via questionnaires and focus groups with students and teaching staff to allow for identification of benefits and challenges. Findings indicated a positive correlation between study engagement and engagement with peers. Students reported that the chatbot enabled them to obtain support and connect to their programme leader. Both staff and students also made recommendation on how engagement could be further enhanced using the bot in terms of; clearly specified purpose, integration with existing university systems, leading by example and connectivity. Extending upon these recommendations, a second pilot study is planned for September 2020, for which the focus will be upon improving attendance rates, student satisfaction and module pass rates.

Competitors’ Influence Analysis of a Retailer by Using Customer Value and Huff’s Gravity Model

Customer relationship analysis is vital for retail stores, especially for supermarkets. The point of sale (POS) systems make it possible to record the daily purchasing behaviors of customers as an identification point of sale (ID-POS) database, which can be used to analyze customer behaviors of a supermarket. The customer value is an indicator based on ID-POS database for detecting the customer loyalty of a store. In general, there are many supermarkets in a city, and other nearby competitor supermarkets significantly affect the customer value of customers of a supermarket. However, it is impossible to get detailed ID-POS databases of competitor supermarkets. This study firstly focused on the customer value and distance between a customer's home and supermarkets in a city, and then constructed the models based on logistic regression analysis to analyze correlations between distance and purchasing behaviors only from a POS database of a supermarket chain. During the modeling process, there are three primary problems existed, including the incomparable problem of customer values, the multicollinearity problem among customer value and distance data, and the number of valid partial regression coefficients. The improved customer value, Huff’s gravity model, and inverse attractiveness frequency are considered to solve these problems. This paper presents three types of models based on these three methods for loyal customer classification and competitors’ influence analysis. In numerical experiments, all types of models are useful for loyal customer classification. The type of model, including all three methods, is the most superior one for evaluating the influence of the other nearby supermarkets on customers' purchasing of a supermarket chain from the viewpoint of valid partial regression coefficients and accuracy.

Minimum Data of a Speech Signal as Special Indicators of Identification in Phonoscopy

Voice biometric data associated with physiological, psychological and other factors are widely used in forensic phonoscopy. There are various methods for identifying and verifying a person by voice. This article explores the minimum speech signal data as individual parameters of a speech signal. Monozygotic twins are believed to be genetically identical. Using the minimum data of the speech signal, we came to the conclusion that the voice imprint of monozygotic twins is individual. According to the conclusion of the experiment, we can conclude that the minimum indicators of the speech signal are more stable and reliable for phonoscopic examinations.