Search

Fuzzy integrals (FIs) are powerful aggregation operators that fuse information from multiple sources. The aggregation is parameterized using a fuzzy measure (FM), which encodes the worths of all subsets of sources. Since the FI is defined with respect to an FM, much consideration must be given to defining the FM. However, in practice this is a difficult task—the number of values in an FM scales as $2^n$ , where $n$ is the number of input sources, thus manually specifying an FM quickly becomes tedious. In this article, we review an automatic, data-supported method of learning the FM by minimizing a sum-of-squared error objective function in the context of decision-level fusion of classifiers using the Choquet FI. While this solves the specification problem, we illuminate an issue encountered with many real-world data sets; i.e., if the training data do not contain a significant number of all possible sort orders, many of the FM values are not supported by the data. We propose various regularization strategies to alleviate this issue by pushing the learned FM toward a predefined structure; these regularizers allow the user to encode knowledge of the underlying FM to the learning problem. Furthermore, we propose another regularization strategy that constrains the learned FM's structure to be a linear order statistic. Finally, we perform several experiments using synthetic and real-world data sets and show that our proposed extensions can improve the learned FM behavior and classification accuracy. A previously proposed visualization technique is employed to simultaneously quantitatively illustrate the FM as well as the FI.

Zadeh's extension principle (ZEP) is a fundamental concept in fuzzy set (FS) theory that enables crisp mathematical operation on FSs. A well-known shortcoming of ZEP is that the height of the output FS is determined by the lowest height of the input FSs. In this article, we introduce a generalized extension principle (GEP) that eliminates this weakness and provides flexibility and control over how membership values are mapped from input to output. Furthermore, we provide a computationally efficient point-based FS representation. In light of our new definition, we discuss two approaches to perform aggregation of FSs using the Choquet integral. The resultant integrals generalize prior work and lay a foundation for future extensions. Last, we demonstrate the extended integrals via a combination of synthetic and real-world examples.

The modern era of machine learning is focused on data-driven solutions. While this has resulted in astonishing leaps in numerous applications, explainability has not witnessed the same growth. The reality is, most machine learning solutions are black boxes. Herein, we focus on data/information fusion in machine learning. Specifically, we explore four eXplainable Artificial Intelligence (XAI) questions relative to Choquet integral; (i) what is the quality of our inputs and their interactions, (ii) how is the information being combined, (iii) what is the quality of our training data (and thus our learned models), and (iv) what trust do we place in an output? Previously, we derived an initial set of indices for (i)–(iv) on the premise of perfect knowledge. Herein, we make XAI more accurate by taking into consideration what the machine learned. A combination of synthetic data and real-world experiments from remote sensing for fusing deep learners in the context of classification are explored. Our approach leads to performance gain, insights into what was learned, and it helps us realize better future solutions.

Mathematical morphology has been explored in deep learning architectures, as a substitute to convolution, for problems like pattern recognition and object detection. One major advantage of using morphology in deep learning is the utility of morphological erosion and dilation. Specifically, these operations naturally embody interpretability due to their underlying connections to the analysis of geometric structures. While the use of these operations results in explainable learned filters, morphological deep learning lacks attribution modeling, i.e., a paradigm to specify what areas of the original observed image are important. Furthermore, convolution-based deep learning has achieved attribution modeling through a variety of neural eXplainable Artificial Intelligence (XAI) paradigms (e.g., saliency maps, integrated gradients, guided backpropagation, and gradient class activation mapping). Thus, a problem for morphology-based deep learning is that these XAI methods do not have a morphological interpretation due to the differences in the underlying mathematics. Herein, we extend the neural XAI paradigm of saliency maps to morphological deep learning, and by doing, so provide an example of morphological attribution modeling. Furthermore, our qualitative results highlight some advantages of using morphological attribution modeling.

Essential oils are mostly used in aromatherapy and their popularity has grown rapidly for the last decade. However, the industry has a substantial low-value waste stream, because downstream industries require therapeutic-grade oil. These waste stream oils can be used in the transport and agricultural sectors. This study investigated the influence of various essential oil blends on the emission characteristics of a multi-cylinder diesel engine. Orange, eucalyptus and tea tree oil were blended with diesel at 5% and 10% by volume, neat diesel and a 10% waste cooking biodiesel-diesel blend were also tested for comparison. The major constituents of orange oil and eucalyptus oil are limonene and 1,8-cineole respectively, and the main constituents of tea tree oil are terpinen-4-ol, γ-terpinene and α-terpinene. Orange oil contains negligible amounts of oxygen, whereas eucalyptus oil and tea tree oil contain 8.4% and 5.4% respectively. Compared to neat diesel, all the essential oil blends exhibited similar or slightly higher density, similar heating value, lower viscosity, flash point, and cetane number, and higher surface tension. However, only orange oil and eucalyptus oil blends exhibited oxidation stability above the minimum standards. Interestingly, blending eucalyptus oil increased the oxidation stability of diesel. Tea tree oil blends emitted the most carbon monoxide (CO) while orange oil and eucalyptus oil blends emitted the least CO and nitrogen oxide (NOX). Although eucalyptus oil and tea tree oil blends contain similar levels of oxygen, they exhibited opposite NOX emission trends, which might be attributed to the dissimilar and complex bonding of oxygen molecules into the structures. Particle number emission of essential oils were load dependent, however, all essential oil belnds emitted higher particulate mass at all loads.

Steganography is the art of hiding secret data in a medium whereas visual-cryptography is a cryptographic technique where a full image is encrypted. The security of image steganography depends on the secret key set which is used for embedding the secret image. If the secret key is exposed, the secret image can be retrieved. To solve this security issue we propose a new type of steganography method using visual cryptography. In this method, text and image are used as a secret message and image for the cover object. 24-bit RGB color images are used as both secret and cover images. In this method, we use a new image namely share1 which converts a secret image to a totally different image called share2 image. Typically, knowing the extraction method, people can retrieve the secret message easily. A pseudorandomly generated image is used as a key for visual encryption and this ensures the extra layer of security. By this proposed method, we ensure more secured image steganography.

Steganography is the art of hiding secret data in a medium. The medium can be an image, video, or an audio file, which is intended to raise no suspicion. Steganography tools typically generate an identical cover image after concealing the secret data in the original image. In this paper, popular steganography tools that are implemented with different embedding algorithms have been compared based on empirical results. Firstly, a set of diverse cover and secret images has been selected as an input set. A number of image quality metrics are applied to the resulting images for comparison. It is found that OpenStego and S-Tools perform the best among seven steganography tools. This comparative analysis gives an insight into the problems and flaws of those tools. Implications for future research and software development for enhancing steganography tools are also discussed based on the experimentation results.

Deep convolutional neural networks (DCNN) have been the dominant methodology in the field of computer vision over the last decade, using various architectural organizations of successive convolutional layers to extract and assemble low level image features into visual component detectors. One of the tradeoffs that have been made as the community has migrated to deep neural models is the loss of explainability and understanding of which salient visual components are being recognized by a model for a particular task. However, there exists a significant heritage in the remote sensing community that has developed advanced algorithms to analyze the signal and structural characteristics of anthropogenic features. One such approach is the use of morphological image processing techniques to extract objects from imagery and aid in the structural analysis of shapes. In particular, the differential morphological profile (DMP) has had great success extracting object shapes, while naturally grouping the extracted shapes into scale ranges. In this research, we present a novel architecture that integrates an explicit (definable and explainable) scaled object extraction into the network architecture, allowing shallower convolutional layers and lower complexity neural models. The architecture is evaluated on a challenging remote sensing dataset of object classes, providing insights to this approach and illuminating future directions of integrating morphology into neural architectures for enhanced explainability.

Fuzzy integrals are useful general purpose aggregation operators, but they can be difficult to understand and visualize in practice. The interaction between an exponentially increasing number of variables–2n fuzzy measure variables for n inputs–makes it hard to understand what exactly is going on in a high dimensional space. We propose a new visualization scheme based on a weighted indicator matrix to better understand the inner workings of an arbitrary fuzzy measure. We provide ways of viewing the Shapley and interaction indices, as well as an optional data coverage histogram. This approach can give insight into which sources are the most relevant in the overall aggregation and decision making process, and it provides a way to visually compare fuzzy measures and subsequently integrals.

Artificial neural networks are a dominant force in our modern era of data-driven artificial intelligence. The adaptive neuro fuzzy inference system (ANFIS) is a neural network based on fuzzy logic versus a more traditional premise like convolution. Advantages of ANFIS include the ability to encode and potentially understand machine learned neural information in the pursuit of explainable, interpretable, and ultimately trustworthy artificial intelligence. However, real-world data is almost always imperfect, e.g., incomplete or noisy, and ANFIS is not naturally robust. Specifically, ANFIS is susceptible to over inflated uncertainty, poor antecedent (fuzzy set) data alignment, degenerate optimization conditions, and hard to interpret logic, to name a few factors. Herein, we explore the use of possibilistic clustering to identify outliers, specifically typicality degrees, to increase the robustness of ANFIS; or any fuzzy logic neuron/network. Experiments are presented that demonstrate the need and quality of the proposed solutions in the pursuit of robust interpretable machine learned neuro fuzzy logic solutions.

The Choquet integral (ChI) is a parametric nonlinear aggregation function defined with respect to the fuzzy measure (FM). To date, application of the ChI has sadly been restricted to problems with relatively few numbers of inputs; primarily as the FM has $2^N$ variables for $N$ inputs and $N(2^{N-1}-1)$ monotonicity constraints. In return, the community has turned to density-based imputation (e.g., Sugeno $\lambda$ -FM) or the number of interactions (FM variables) are restricted (e.g., $k$ -additivity). Herein, we propose a new scalable data-driven way to represent and learn the ChI, making learning computationally manageable for larger $N$ . First, data supported variables are identified and used in optimization. Identification of these variables also allows us recognize future ill-posed fusion scenarios; ChIs involving variable subsets not supported by data. Second, we outline an imputation function framework to address data unsupported variables. Third, we present a lossless way to compress redundant variables and associated monotonicity constraints. Finally, we outline a lossy approximation method to further compress the ChI (if/when desired). Computational complexity analysis and experiments conducted on synthetic datasets with known FMs demonstrate the effectiveness and efficiency of the proposed theory.

Waste management cost for Australia is increasing every year, and thus, it is important to find alternative ways to use the waste. For example, essential oil has a significant waste stream that can be utilized in vehicles of their producers. However, some of the essential oils contain oxygen which considerably affects engine performance, emission, and combustion characteristics of diesel engines. Thus, this research paper will try to evaluate the essential oils as a replacement of diesel fuel to operate a multicylider diesel engine. For this study, two essential oils are selected which contain different oxygenated functional groups, tea tree oil (5.4% oxygen) and eucalyptus oil (8.4% oxygen), with an aim to evaluate the effect of these functional groups on engine performance and emission parameters. These oils were blended with neat diesel (0% oxygen) to obtain a blend cotaining 2.2% oxygen by weight. The blends produced similar brake power; however, brake-specific fuel consumption (BSFC) increased for eucalyptus oil blends (2.4–3.7%) and tea tree oil blends (3.9–5.3%). Essential oil–diesel blends resulted in less CO and increased NOX emission, produced similar peak pressure, and indicated mean effective pressure. The results then lead to the conclusion that oxygenated essential oils can have a role to reduce dependency of agricultural sector on diesel in the near future.

This study explored the activity of moving vehicle and public transportation, as a next stage of smart device use, and its evolution with data science. Smart devices have been enabled to help people carry out their daily activities. Some smart device operating systems have special apps to make it easy for users to record information about walking, running, jogging, and stepping through the Activity Recognition process, even recording physiological information. Advanced smart devices can now be used to record information about moving vehicles, as it has begun by enabling paid advertising on personal motor vehicles. However, with advancements in smart device technology, as analyzed and proposed by this paper, people will be able to utilize this technology with new data science methods to analyze far more information related to moving vehicles and public transportation. In this paper, we have proposed a method to recognize the vehicle that the subject is currently using by recorded accelerometer and gyroscope sensor data embedded in a smart device. In other words, we have implemented a machine learning model Artificial Neural Network to classify vehicles from the sensor data.

Microalgae feedstock is recognised as one of the most promising resources for producing triglycerides which is subsequently converted to biodiesel. However, the large-scale technology required to generate biodiesel from microalgae is still in its early stages of development. Microalgae research to date may be placed into four broad categories: (i) growth, (ii) harvesting, (iii) oil extraction and (iv) fuel properties for engine performance and emissions. More than 1000 manuscripts have been published on the first category with progressively less on subsequent groups. Finally, effects of microalgae methyl esters on engine performance have only been reported in 9 scientific articles. This review will place extraction techniques and engine performance of microalgae biodiesel in the context of the preceding two categories and examine the practical problems associated with fuel properties, engine performance and emissions. Considering energy consumption, toxicity, and time, many of the extraction techniques used in the laboratory show moderate potential for commercial scale. An important finding is that variation of conditions in the first three categories can have a significant effect on biofuel quality which can cause fuel properties to be out of standard and/or adversely affect engine performance and emissions.

Since the state-of-the-art deep learning algorithms demand a large training dataset, which is often unavailable in some domains, the transfer of knowledge from one domain to another has been a trending technique in the computer vision field. However, this method may not be a straight-forward task considering several issues such as original network size or large differences between the source and target domain. In this paper, we perform transfer learning for semantic segmentation of off-road driving environments using a pre-trained segmentation network called DeconvNet. We explore and verify two important aspects regarding transfer learning. First, since the original network size was very large and did not perform well for our application, we proposed a smaller network, which we call the light-weight network. This light-weight network is half the size to the original DeconvNet architecture. We transferred the knowledge from the pre-trained DeconvNet to our light-weight network and fine-tuned it. Second, we used synthetic datasets as the intermediate domain before training with the real-world off-road driving data. Fine-tuning the model trained with the synthetic dataset that simulates the off-road driving environment provides more accurate results for the segmentation of real-world off-road driving environments than transfer learning without using a synthetic dataset does, as long as the synthetic dataset is generated considering real-world variations. We also explore the issue whereby the use of a too simple and/or too random synthetic dataset results in negative transfer. We consider the Freiburg Forest dataset as a real-world off-road driving dataset.

Herein, a generalization of the ordered weighted average (OWA) is put forth relative to pattern recognition. The resultant linear order statistic neuron (LOSN) is unique in that it bridges fuzzy sets, specifically fuzzy data/information aggregation, with neural networks. This article discusses the gradient descent-based optimization and geometric interpretation of the LOSN. An advantage is that the LOSN is an efficient shared weight encoding of N! perceptrons, relative to N inputs. Open source codes are provided to facilitate reproducible research. Experiments are conducted to both validate the method and show its non-linear geometric expression.

The Choquet integral (ChI) is a proven tool for information aggregation. In prior work, we showed that learning a ChI from data results in missing variables. Herein, we explore two ways to transfer a known ChI from a source domain to a new under sampled target domain. The first method is based on regularization and it listens to the full source domain ChI. The second method optimizes what we can observe (target domain supported variables) and missing variables are the only thing migrated from the source domain. Synthetic experiments, aka we know the truth, are used to show the behavior of these methods with respect to transfering between ChIs.

Information fusion is an essential part of numerous engineering systems and biological functions, e.g., human cognition. Fusion occurs at many levels, ranging from the low-level combination of signals to the high-level aggregation of heterogeneous decision-making processes. While the last decade has witnessed an explosion of research in deep learning, fusion in neural networks has not observed the same revolution. Specifically, most neural fusion approaches are ad hoc, are not understood, are distributed versus localized, and/or explainability is low (if present at all). Herein, we prove that the fuzzy Choquet integral (ChI), a powerful nonlinear aggregation function, can be represented as a multi-layer network, referred to hereafter as ChIMP. We also put forth an improved ChIMP (iChIMP) that leads to a stochastic gradient descent-based optimization in light of the exponential number of ChI inequality constraints. An additional benefit of ChIMP/iChIMP is that it enables eXplainable AI (XAI). Synthetic validation experiments are provided and iChIMP is applied to the fusion of a set of heterogeneous architecture deep models in remote sensing. We show an improvement in model accuracy and our previously established XAI indices shed light on the quality of our data, model, and its decisions., IEEE Transactions on Fuzzy Systems

For autonomous vehicles 3D, rotating LiDAR sensors are often critically important towards the vehicle’s ability to sense its environment. Generally, these sensors scan their environment, using multiple laser beams to gather information about the range and the intensity of the reflection from an object. LiDAR capabilities have evolved such that some autonomous systems employ multiple rotating LiDARs to gather greater amounts of data regarding the vehicle’s surroundings. For these multi–LiDAR systems, the placement of the sensors determine the density of the combined point cloud. We perform preliminary research regarding the optimal LiDAR placement strategy on an off–road, autonomous vehicle known as the Halo project. We use the Mississippi State University Autonomous Vehicle Simulator (MAVS) to generate large amounts of labeled LiDAR data that can be used to train and evaluate a neural network used to process LiDAR data in the vehicle. The trained networks are evaluated and their performance metrics are then used to generalize the performance of the sensor pose. Data generation, training, and evaluation, was performed iteratively to perform a parametric analysis of the effectiveness of various LiDAR poses in the Multi–LiDAR system. We also, describe and evaluate intrinsic and extrinsic calibration methods that are applied in the multi–LiDAR system. In conclusion we found that our simulations are an effective way to evaluate the efficacy of various LiDAR placements based on the performance of the neural network used to process that data and the density of the point cloud in areas of interest.

Essential oils are obtained from the non-fatty parts of a plant, such as the roots, bark, leaves, stems and flowers. These oils are mainly used in the natural medicine sector due to claimed health benefits, as well as the flavouring and fragrance sector, and the market has experienced rapid growth in recent years. The high quality required of the products leads to a very significant low-value waste stream, which is available for use in the transport and agricultural sectors. The use of essential oils in the compression ignition (CI) engine is a concept that has not yet been explored thoroughly. This paper analyses the available literature on the effect of essential oils and their blends on the performance, combustion characteristics and emission parameters of the CI engine. Regarding their properties, essential oils have similar properties to neat diesel. Engine performance using several essential oils and their blends improve brake thermal efficiency (BTE) and reduce brake-specific fuel consumption. A significant reduction in particulate matter (PM) emissions along with reduced hydrocarbon and carbon monoxide emissions has been reported in the literature. Furthermore, essential oils and their blends increase peak cylinder pressure and heat release rate (HRR) compared to neat diesel and biodiesel. However, most of the literature reported increased emissions of nitrogen oxides attributed to the lower cetane number and higher oxygen content. Due to having a low cetane number, essential oils can be used in CI engines by blending with either diesel or biodiesel. The rapid growth of the essential oil sector increases the likelihood of their utilization in CI engines in the future.