Health technological systems learning from and reacting on how humans behave in sensor equipped environments are today being commercialized. These systems rely on the assumptions that training data and testing data share the same feature space, and residing from the same underlying distribution - which is commonly unrealistic in real-world applications. Instead, the use of transfer learning could be considered. In order to transfer knowledge between a source and a target domain these should be mapped to a common latent feature space. In this work, the dimensionality reduction algorithm t-SNE is used to map data to a similar feature space and is further investigated through a proposed novel analysis of output stability. The proposed analysis, Normalized Linear Procrustes Analysis (NLPA) extends the existing Procrustes and Local Procrustes algorithms for aligning manifolds. The methods are tested on data reflecting human behaviour patterns from data collected in a smart home environment. Results show high partial output stability for the t-SNE algorithm for the tested input data for which NLPA is able to detect clusters which are individually aligned and compared. The results highlight the importance of understanding output stability before incorporating dimensionality reduction algorithms into further computation, e.g. for transfer learning.

Human activity recognition as an engineering tool as well as an active research field has become fundamental to many applications in various fields such as health care, smart home monitoring and surveillance. However, delivering sufficiently robust activity recognition systems from sensor data recorded in a smart home setting is a challenging task. Moreover, human activity datasets are typically highly imbalanced because generally certain activities occur more frequently than others. Consequently, it is challenging to train classifiers from imbalanced human activity datasets. Deep learning algorithms perform well on balanced datasets, yet their performance cannot be promised on imbalanced datasets. Therefore, we aim to address the problem of class imbalance in deep learning for smart home data. We assess it with Activities of Daily Living recognition using binary sensors dataset. This paper proposes a data level perspective combined with a temporal window technique to handle imbalanced human activities from smart homes in order to make the learning algorithms more sensitive to the minority class. The experimental results indicate that handling imbalanced human activities from the data-level outperforms algorithms level and improved the classification performance. © The Author(s) 2020

A demographic change is occurring in many areas of the world. The elderly population share has been increasing for the last decades and estimations predict that this group will be large in proportion to the number of economically active younger people. This change will bring exponentially increasing costs of health care. Technical developments could be one way to meet these new challenges. In a recent study called “Safe at night” the aim was to investigate whether a technical solution based can be used to supplement the home care work, with focus on the nightly visits of the elderly. The study raises questions regarding technical issues as well as actors (users, relatives and staffs) perspective on the methods. Researchers from both social and technical disciplines were involved in the study. In this paper, we highlight the importance of scientists from different disciplines participating in the study, as well as municipalities and industry. We show in particular the knowledge gained from a technical perspective and from a social science perspective and how and why these perspectives together constitute the necessary components to create innovation regarding elderly care and issues related to technology and trust.

Data anonymization has been used as a fundamental tool in various domains, e.g. healthcare, to alter personal data such that individuals can no longer be identified directly or indirectly in a way to enable broader sharing of data. For example, data perturbation techniques add noise to original data allowing individual record confidentiality while maintaining high-quality data for analytical purposes. In this paper, we propose a perturbation technique for anonymizing longitudinal tabular data such as electronic health records (EHRs). Our model starts by learning a latent space of original data to better capture temporal trends, then employs a generative adversarial network together to train a perturbation generator. During model training, a time-supervised loss function for handling sequence-dependent noise, together with the adversarial unsupervised, anonymization, and reconstruction loss functions are utilized. To evaluate our model quantitatively, we use multiple evaluation metrics for the fidelity, utility, and identifiability of generated data, in addition, the model is evaluated qualitatively by visualizing generated and original data. The results confirm that our model preserves the privacy of the original data and generates a perturbed version with high fidelity and utility compared to some state-of-the-art techniques. © 2023 IEEE.

Healthcare longitudinal data collected around patients' life cycles, today offer a multitude of opportunities for healthcare transformation utilizing artificial intelligence algorithms. However, access to "real" healthcare data is a big challenge due to ethical and legal reasons. There is also a need to deal with challenges around electronic health records (EHRs) including biased, heterogeneity, imbalanced data, and small sample sizes. In this study, we introduce a domain knowledge-driven framework for generating synthetic EHRs, as an alternative to methods only using EHR data or expert knowledge. By leveraging external medical knowledge sources in the training algorithm, the suggested framework is designed to maintain data utility, fidelity, and clinical validity while preserving patient privacy. © 2023 European Federation for Medical Informatics (EFMI) and IOS Press.

The purpose of the study is to demonstrate the feasibility of combining traffic simulator technology with machine learning (ML) methods to create realistic and comprehensive synthetic traffic data. Synthetic data alleviates many ethical and privacy concerns, significantly reduces the costs associated with data collection, and enables researchers to study scenarios and conditions that are difficult or impossible to replicate in real-world environments. Access to large amounts of diverse and controlled data is essential for developing and testing artificial intelligence (AI) models and leads to more reliable and robust results. Traffic simulators like SUMO have been successfully used for that purpose in the past, creating realistic vehicular traces. One drawback is that, without coupling them with complex physics emulators, they are not capable of generating internal vehicle parameters. Such parameters, on the other hand, are crucial for many purposes, from understanding energy efficiency and optimizing driver behavior to predictive maintenance and monitoring the degradation of key components, such as driveline batteries. In this paper, we propose Synthetic Traffic Data Generator (STDG) and demonstrate that an ML model that is trained on the internal parameters of a vehicle in one set of conditions (Sweden) can be used to generate synthetic data corresponding to another setting (Monaco). The proposed method promises to eliminate the need for an expensive collection of the original vehicle parameters across many different settings. Moreover, sharing the synthetic data with additional stakeholders is easier due to the reduced security and integrity risk of exposing the vehicle’s privacy-sensitive original parameters. This study compares several ML techniques, including deep learning (DL) based, for generating internal parameters of vehicles, such as fuel rate, engine speed, and wet tank air pressure. Using the actual bus data from a small city to train our ML models, we attempt to forecast the internal parameters of the buses in various scenarios. The proposed method first utilizes SUMO to generate synthetic waypoints for the bus and then predicts the other parameters using the trained model, thereby producing synthetic data with internal parameters for buses operating in a new urban environment. Our preliminary results indicated that our model is performing well within a 90% confidence interval. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Vehicle utilization analysis is an essential tool for manufacturers to understand customer needs, improve equipment uptime, and to collect information for future vehicle and service development. Typically today, this behavioral modeling is done on high-resolution time-resolved data with features such as GPS position and fuel consumption. However, high-resolution data is costly to transfer and sensitive from a privacy perspective. Therefore, such data is typically only collected when the customer pays for extra services relying on that data. This motivated us to develop a multi-task ensemble approach to transfer knowledge from the high-resolution data and enable vehicle behavior prediction from low-resolution but high dimensional data that is aggregated over time in the vehicles.

This study proposes a multi-task snapshot-stacked ensemble (MTSSE) deep neural network for vehicle behavior prediction by considering vehicles’ low-resolution operational life records. The multi-task ensemble approach utilizes the measurements to map the low-frequency vehicle usage to the vehicle behaviors defined from the high-resolution time-resolved data. Two data sources are integrated and used: high-resolution data called Dynafleet, and low-resolution so-called Logged Vehicle Data (LVD). The experimental results demonstrate the proposed approach’s effectiveness in predicting the vehicle behavior from low frequency data. With the suggested multi-task snapshot-stacked ensemble deep network, it is shown how low-resolution sensor data can highly contribute to predicting multiple vehicle behaviors simultaneously while using only one single training process. © 2022 The Author(s)

Predicting breakdowns is becoming one of the main goals for vehicle manufacturers so as to better allocate resources, and to reduce costs and safety issues. At the core of the utilization of vehicle sensors is the fact that early detection of anomalies facilitates the prediction of potential breakdown issues, which, if otherwise undetected, could lead to breakdowns and warranty claims. However, the making of such predictions is too complex a challenge to solve using simple predictive models. The strength of heuristic optimization techniques in solving np-hard problems, and the recent success of ensemble approaches to various modeling problems, motivated us to investigate a hybrid optimization- and ensemble-based approach to tackle the complex task. In this study, we propose a snapshot-stacked ensemble deep neural network (SSED) approach to predict vehicle claims (in this study, we refer to a claim as being a breakdown or a fault) by considering vehicle operational life records. The approach includes three main modules: Data pre-processing, Dimensionality Reduction, and Ensemble Learning. The first module is developed to run a set of practices to integrate various sources of data, extract hidden information and segment the data into different time windows. In the second module, the most informative measurements to represent vehicle usage are selected through an adapted heuristic optimization approach. Finally, in the last module, the ensemble machine learning approach utilizes the selected measurements to map the vehicle usage to the breakdowns for the prediction. The proposed approach integrates, and uses, the following two sources of data, collected from thousands of heavy-duty trucks: Logged Vehicle Data (LVD) and Warranty Claim Data (WCD). The experimental results confirm the proposed system’s effectiveness in predicting vehicle breakdowns. By adapting the optimization and snapshot-stacked ensemble deep networks, we demonstrate how sensor data, in the form of vehicle usage history, contributes to claim predictions. The experimental evaluation of the system on other application domains also indicated the generality of the proposed approach. © 2023 by the authors.

Machine learning methods are increasingly being used to solve real-world problems in the society. Often, the complexity of the methods are well hidden for users. However, integrating machine learning methods in real-world applications is not a straightforward process and requires knowledge both about the methods and domain knowledge of the problem. Two such domains are colour print quality assessment and anomaly detection in smart homes, which are currently driven by manual monitoring of complex situations. The goal of the presented work is to develop methods, algorithms and tools to facilitate monitoring and understanding of the complex situations which arise in colour print quality assessment and anomaly detection for smart homes. The proposed approach builds on the use and adaption of supervised and unsupervised machine learning methods.

Novel algorithms for computing objective measures of print quality in production are proposed in this work. Objective measures are also modelled to study how paper and press parameters influence print quality. Moreover, a study on how print quality is perceived by humans is presented and experiments aiming to understand how subjective assessments of print quality relate to objective measurements are explained. The obtained results show that the objective measures reflect important aspects of print quality, these measures are also modelled with reasonable accuracy using paper and press parameters. The models of objective  measures are shown to reveal relationships consistent to known print quality phenomena.

In the second part of this thesis the application area of anomaly detection in smart homes is explored. A method for modelling human behaviour patterns is proposed. The model is used in order to detect deviating behaviour patterns using contextual information from both time and space. The proposed behaviour pattern model is tested using simulated data and is shown to be suitable given four types of scenarios.

The thesis shows that parts of offset lithographic printing, which traditionally is a human-centered process, can be automated by the introduction of image processing and machine learning methods. Moreover, it is concluded that in order to facilitate robust and accurate anomaly detection in smart homes, a holistic approach which makes use of several contextual aspects is required.

Understanding the aspects of progression for atherosclerotic cardiovascular disease and treatment is key to building reliable clinical decision-support systems. To promote system trust, one step is to make the machine learning models (used by the decision support systems) explainable for clinicians, developers, and researchers. Recently, working with longitudinal clinical trajectories using Graph Neural Networks (GNNs) has attracted attention among machine learning researchers. Although GNNs are seen as black-box methods, promising explainable AI (XAI) methods for GNNs have lately been proposed. In this paper, which describes initial project stages, we aim at utilizing GNNs for modeling, predicting, and exploring the model explainability of the low-density lipoprotein cholesterol level in long-term atherosclerotic cardiovascular disease progression and treatment.

A system for detecting deviating human behaviour in a smart home environment is the long-term goal of this work. It is believed that such systems will be very important in ambient assisted living services. Three types of deviations are considered in this work: deviation in activity intensity, deviation in time and deviation in space. Detection of deviations in activity intensity is formulated as the on-line quickest detection of a parameter shift in a sequence of independent Poisson random variables. Random forests trained in an unsupervised fashion are used to learn the spatial and temporal structure of data representing normal behaviour and are thereafter utilised to find deviations.The experimental investigations have shown that the Page and Shiryaev change-point detection methods are preferable in terms of expected delay of motivated alarm. Interestingly only a little is lost when the methods are specified with estimated intensity parameters rather than the true intensity values which are not available in a real situation. As to the spatial and temporal deviations, they can be revealed through analysis of a 2D map of high dimensional data. It was demonstrated that such a map is stable in terms of the number of clusters formed. We have shown that the data clusters can be understood/explored by finding the most important variables and by analysing the structure of the most representative tree.

A system for detecting deviating human behaviour in a smart home environment is the long-term goal of this work. Clearly, such systems will be very important in ambient assisted living services. A new approach to modelling human behaviour patterns is suggested in this paper. The approach reveals promising results in unsupervised modelling of human behaviour and detection of deviations by using such a model. Human behaviour/activity in a short time interval is represented in a novel fashion by responses of simple non-intrusive sensors. Deviating behaviour is revealed through data clustering and analysis of associations between clusters and data vectors representing adjacent time intervals (analysing transitions between clusters). To obtain clusters of human behaviour patterns, first, a random forest is trained without using beforehand defined teacher signals. Then information collected in the random forest data proximity matrix is mapped onto the 2D space and data clusters are revealed there by agglomerative clustering. Transitions between clusters are modelled by the third order Markov chain.

Three types of deviations are considered: deviation in time, deviation in space and deviation in the transition between clusters of similar behaviour patterns.

The proposed modelling approach does not make any assumptions about the position, type, and relationship of sensors but is nevertheless able to successfully create and use a model for deviation detection-this is claimed as a significant result in the area of expert and intelligent systems. Results show that spatial and temporal deviations can be revealed through analysis of a 2D map of high dimensional data. It is demonstrated that such a map is stable in terms of the number of clusters formed. We show that the data clusters can be understood/explored by finding the most important variables and by analysing the structure of the most representative tree. © 2016 Elsevier Ltd. All rights reserved.

Applying machine learning methods in scenarios involving smart homes is a complex task. The many possible variations of sensors, feature representations, machine learning algorithms, middle-ware architectures, reasoning/decision schemes, and interactive strategies make research and development tasks non-trivial to solve.In this paper, the use of a portable, flexible and holistic smart home demonstrator is proposed to facilitate iterative development and the acquisition of feedback when testing in regard to the above-mentioned issues. Specifically, the focus in this paper is on scenarios involving anomaly detection and response. First a model for anomaly detection is trained with simulated data representing a priori knowledge pertaining to a person living in an apartment. Then a reasoning mechanism uses the trained model to infer and plan a reaction to deviating activities. Reactions are carried out by a mobile interactive robot to investigate if a detected anomaly constitutes a true emergency. The implemented demonstrator was able to detect and respond properly in 18 of 20 trials featuring normal and deviating activity patterns, suggesting the feasibility of the proposed approach for such scenarios. © IEEE 2015

Research on intelligent environments, such as smart homes, concerns the mechanisms that intelligently orchestrate the pervasive technical infrastructure in the environment. However, significant challenges are to build, configure, use and maintain these systems. Providing personalized services while preserving the privacy of the occupants is also difficult. As an approach to facilitate research in this area, this paper presents the Halmstad Intelligent Home and a novel approach for multioccupancy detection utilizing the presented environment. This paper also presents initial results and ongoing work. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2016.

Development, testing and validation of algorithms for smart home applications are often complex, expensive and tedious processes. Research on simulation of resident activity patterns in Smart Homes is an active research area and facilitates development of algorithms of smart home applications. However, the simulation of passive infrared (PIR) sensors is often used in a static fashion by generating equidistant events while an intended occupant is within sensor proximity. This paper suggests the combination of avatar-based control and probabilistic sampling in order to increase realism of the simulated data. The number of PIR events during a time interval is assumed to be Poisson distributed and this assumption is used in the simulation of Smart Home data. Results suggest that the proposed approach increase realism of simulated data, however results also indicate that improvements could be achieved using the geometric distribution as a model for the number of PIR events during a time interval. © IEEE 2015

Information processing steps in printing industry are highly automated, except the last one print quality assessment, which usually is a manual, tedious, and subjective procedure. This article presents a random forests-based technique for automatic print quality assessment based on objective values of several printquality attributes. Values of the attributes are obtained from soft sensors through data mining and colour image analysis. Experimental investigations have shown good correspondence between print quality evaluations obtained by the technique proposed and the average observer. (C) 2012 Elsevier B.V. All rights reserved.

Nowadays in printing industry most of information processing steps are highly automated, except the last one–print quality assessment and control. We present a way to assess one important aspect of print quality, namely the distortion of halftone dots printed colour pictures are made of. The problem is formulated as assessing the distortion of circles detected in microscale images of halftone dot areas. In this paper several known circle detection techniques are explored in terms of accuracy and robustness. We also present a new circle detection technique based on the fuzzy Hough transform (FHT) extended with k-means clustering for detecting positions of accumulator peaks and with an optional fine-tuning step implemented through unsupervised learning. Prior knowledge about the approximate positions and radii of the circles is utilized in the algorithm. Compared to FHT the proposed technique is shown to increase the estimation accuracy of the position and size of detected circles. The techniques are investigated using synthetic and natural images.

Variations in offset print quality relate to numerous parameter of printing press and paper. To maintain constant quality of products, press operators need to assess, explore and monitor print quality. This paper presents a novel system for assessing and predicting values of print quality attributes, where the adopted, random forests (RF)-based, modeling approach also allows quantifying the influence of different parameters. In contrast to other print quality assessment systems, this system utilizes common print marks known as double grey-bars. A novel virtual sensor for assessing the mis-registration degree of printing plates using images of double grey-bars is presented. The inferred influence of paper and printing press parameters on print quality shows correlation with known print quality conditions.

Variations in offset print quality relate to numerous parameters of printing press and paper. To maintain a constant high print quality press operators need to assess, explore and monitor quality of prints. Today assessment is mainly done manually. This paper presents a novel system for assessing and predicting values of print quality attributes, where the adopted, random forests (RFs)-based, modeling approach also allows quantifying the influence of different paper and press parameters on print quality. In contrast to other print quality assessment systems the proposed system utilises common, simple print marks known as double grey-bars. Novel virtual sensors assessing print quality attributes using images of double grey-bars are presented. The inferred influence of paper and printing press parameters on quality of colour prints shows clear relation with known print quality conditions. Thorough analysis and categorisation of related work is also given in the paper. (C) 2012 Elsevier Ltd. All rights reserved.

It is fully appreciated that progress in the development of data driven approaches to activity recognition are being hampered due to the lack of large scale, high quality, annotated data sets. In an effort to address this the Open Data Initiative (ODI) was conceived as a potential solution for the creation of shared resources for the collection and sharing of open data sets. As part of this process, an analysis was undertaken of datasets collected using a smart environment simulation tool. A noticeable difference was found in the first 1-2 cycles of users generating data. Further analysis demonstrated the effects that this had on the development of activity recognition models with a decrease of performance for both support vector machine and decision tree based classifiers. The outcome of the study has led to the production of a strategy to ensure an initial training phase is considered prior to full scale collection of the data.

Traditionally, database management systems (DBMSs) have been employed exclusively for data management in infrastructures supporting Ambient Assisted Living (AAL) systems. However, DBMSs provide other mechanisms, such as for security, dependability, and extensibility that can facilitate the development, use, and maintenance of AAL applications. This work utilizes such mechanisms, particularly extensibility, and proposes a database-centric architecture to support home-based healthcare applications. An active database is used to monitor and respond to events taking place in the home, such as bed-exits. In-database data mining methods are applied to model early night behaviors of people living alone. Encapsulating the processing into the DBMS avoids transferring and processing sensitive data outside of database, enables changes in the logic to be managed on-the-fly, and reduces code duplication. As a result, such an approach leads to better performance and increased security and privacy, and can facilitate the adaptability and scalability of AAL systems. An evaluation of the architecture with datasets collected in real homes demonstrated the feasibility and flexibility of the approach.

As an alternative to the existing software architectures that underpin the development of smart homes and ambient assisted living (AAL) systems, this work presents a database-centric architecture that takes advantage of active databases and in-database processing. Current platforms supporting AAL systems use database management systems (DBMSs) exclusively for data storage. Active databases employ database triggers to detect and react to events taking place inside or outside of the database. DBMSs can be extended with stored procedures and functions that enable in-database processing. This means that the data processing is integrated and performed within the DBMS. The feasibility and flexibility of the proposed approach were demonstrated with the implementation of three distinct AAL services. The active database was used to detect bed-exits and to discover common room transitions and deviations during the night. In-database machine learning methods were used to model early night behaviors. Consequently, active in-database processing avoids transferring sensitive data outside the database, and this improves performance, security and privacy. Furthermore, centralizing the computation into the DBMS facilitates code reuse, adaptation and maintenance. These are important system properties that take into account the evolving heterogeneity of users, their needs and the devices that are characteristic of smart homes and AAL systems. Therefore, DBMSs can provide capabilities to address requirements for scalability, security, privacy, dependability and personalization in applications of smart environments in healthcare.

Background: Older adults often want to stay in a familiar place, such as their home, as they get older. This so-called aging in place, which may involve support from relatives or care professionals, can promote older people’s independence and well-being. The combination of aging and disease, however, can lead to complex medication regimes, and difficulties for care providers in correctly assessing the older person's health. In addition, the organization of the health care is fragmented, which makes it difficult for health professionals to encourage older people to participate in their care. It is also a challenge to perform adequate health assessment and appropriate communication between health care professionals.

Objective: The purpose of this paper is to describe the design for an integrated home-based system that can acquire and compile health-related evidence for guidance and information sharing among care providers and care receivers in order to support and promote medication self-management among older people.

Methods: The authors used a participatory design (PD) approach for this mixed-method project, which was divided into four phases: Phase I, Conceptualization, consisted of the conceptualization of a system to support medication self- management, objective health assessments, and communication between health care professionals. Phase II, Development of a System, consisted of building and bringing together the conceptualized systems from phase I. Phases III (pilot study) and IV (a full-scale study) are described briefly.

Results: Our participants in phase I were people who were involved in some way in the care of older adults, and included older adults themselves, relatives of older adults, care professionals, and industrial partners. With input from phase I participants, we identified two relevant concepts for promoting medication self-management, both of which related to systems that participants believed could provide guidance for the older adults themselves, relatives of older adults, and care professionals. The system will also encourage information sharing between care providers and care receivers. The first is the concept of the Intelligent Friendly Home (IAFH), defined as an integrated residential system that evolves to sense, reason and act in response to individual needs, preferences and behaviors as these change over time. The second concept is the MedOP system, a system that would be supported by the IAFH, and which consists of three related components: one that assess health behaviors, another that communicates health data, and a third that promotes medication self-management.

Conclusions: The participants in this project were older adults, relatives of older adults, care professionals, and our industrial partners. With input from the participants, we identified two main concepts that could comprise a system for health assessment, communication and medication self-management: the Intelligent Friendly Home (IAFH), and the MedOP system. These concepts will be tested in this study to determine whether they can facilitate and promote medication self-management in older people. © The authors. All rights reserved. 

This paper addresses approaches to Human Activity Recognition (HAR) with the aim of monitoring the physical activity of people in the workplace, by means of a smartphone application exploiting the available on-board accelerometer sensor. In fact, HAR via a smartphone or wearable sensor can provide important information regarding the level of daily physical activity, especially in situations where a sedentary behavior usually occurs, like inmodern workplace environments. Increased sitting time is significantly associated with severe health diseases, and the workplace is an appropriate intervention setting, due to the sedentary behavior typical of modern jobs. Within this paper, the state-of-the-art components of HAR are analyzed, in order to identify and select the most effective signal filtering and windowing solutions for physical activity monitoring. The classifier development process is based upon three phases; a feature extraction phase, a feature selection phase, and a training phase. In the training phase, a publicly available dataset is used to test among different classifier types and learning methods. A user-friendly Android-based smartphone application with low computational requirements has been developed to run field tests, which allows to easily change the classifier under test, and to collect new datasets ready for use with machine learning APIs. The newly created datasets may include additional information, like the smartphone position, its orientation, and the user's physical characteristics. Using the mobile tool, a classifier based on a decision tree is finally set up and enriched with the introduction of some robustness improvements. The developed approach is capable of classifying six activities, and to distinguish between not active (sitting) and active states, with an accuracy near to 99%. The mobile tool, which is going to be further extended and enriched, will allow for rapid and easy benchmarking of new algorithms based on previously generated data, and on future collected datasets. © 2016 Susanna Spinsante et al.

The development, testing and evaluation of novel approaches to Intelligent Environment data processing require access to datasets which are of high quality, validated and annotated. Access to such datasets is limited due to issues including cost, flexibility, practicality, and a lack of a globally standardized data format. These limitations are detrimental to the progress of research. This paper provides an overview of the Open Data Initiative and the use of simulation software (IE Sim) to provide a platform for the objective assessment and comparison of activity recognition solutions. To demonstrate the approach, a dataset was generated and distributed to 3 international research organizations. Results from this study demonstrate that the approach is capable of providing a platform for benchmarking and comparison of novel approaches.

Recent research has found deep neural networks to be vulnerable, by means of prediction error, to images corrupted by small amounts of non-random noise. These images, known as adversarial examples are created by exploiting the input to output mapping of the network. For the MNIST database, we observe in this paper how well the known regularization/robustness methods improve generalization performance of deep neural networks when classifying adversarial examples and examples perturbed with random noise. We conduct a comparison of these methods with our proposed robustness method, an ensemble of models trained on adversarial examples, able to clearly reduce prediction error. Apart from robustness experiments, human classification accuracy for adversarial examples and examples perturbed with random noise is measured. Obtained human classification accuracy is compared to the accuracy of deep neural networks measured in the same experimental settings. The results indicate, human performance does not suffer from neural network adversarial noise.

Nowadays most of information processing steps in printing industry are highly automated, except the last one – print quality assessment and control. Usually quality assessment is a manual, tedious, and subjective procedure. This article presents a survey of non numerous developments in the field of computational intelligence-based print quality assessment and control in offset colour printing. Recent achievements in this area and advances in applied computational intelligence, expert and decision support systems lay good foundations for creating practical tools to automate the last step of the printing process.