Search

In this paper we present the derivation of two new forms of the Kalman filter equations; the first is for a pure lognormally distributed random variable, while the second set of Kalman filter equations will be for a combination of Gaussian and lognormally distributed random variables. We show that the appearance is similar to that of the Gaussian-based equations, but that the analysis state is a multivariate median and not the mean. We also show results of the mixed distribution Kalman filter with the Lorenz 1963 model with lognormal errors for the background and observations of the z component, and compare them to analysis results from a traditional Gaussian-based extended Kalman filter and show that under certain circumstances the new approach produces more accurate results.

Smart-home systems achieved great popularity in the last decade as they increase the comfort and quality of life. Reduction of energy consumption became a very important desiderate in the context of the explosive technological development of modern society with a major impact on the future development of mankind. Moreover, due to the large amount of data available from smart meters installed in households. It makes leverage to able to find data abnormalities for better monitoring and forecasting. Detecting data anomalies helps in making a better decision for reducing energy usage wasted. In recent years, machine learning models are widely used for developing intelligent systems. Currently, researchers’ main focus is on developing supervised learning models for predicting anomalies. However, there are challenges to train models with unlabeled data indicating data anomaly or not. In this paper, abnormalities are detected in electricity usage using unsupervised learning and evaluated using Excess Mass. The unsupervised anomaly detection model is based on Gaussian Mixture Model (GMM) and Isolation Forest (iForest). The models are compared with Local Outlier Factor (LOF) and One-class support vector machine (OCSVM). The proposed framework is tested with actual electricity usage and temperature data obtained from Numenta Anomaly Benchmark (NAB), which contains normal and anomaly data in time series. Finally, it has been observed that the iForest out-performed as the detection model for the selected use case. The outcome showed that the iForest can quickly detect anomalies in electricity usage data with only a sequence of data without feature extraction. The proposed model is suitable for the Smart Home Energy Management System's practical requirement and can be implemented in various houses independently. The proposed system can also be extended with the various use cases having similar data types.

A job interview can be challenging and stressful even when one has gone through it many times. Failure to handle the stress may lead to unsuccessful delivery of their best throughout the interview session. Therefore, an alternative method which is preparing a video resume and interview before the actual interview could reduce the level of stress. An intelligent stress detection is proposed to classify individuals with different stress levels by understanding the physiological signal through electrocardiogram (ECG) signals. The Augsburg biosignal toolbox (AUBT) dataset was used to obtain the state-of-art results. Only five selected features are significant to the stress level were fed into neural network multi-layer perceptron (MLP) as the optimum classifier. This stress detection achieved an accuracy of 92.93% when tested over the video interview dataset of 10 male subjects who were recording the video resume for the analysis purposes.

Acute lymphoblastic leukemia (ALL) has recently been one of the most significant concerns in cancers, especially child and old age. Therefore, crying needs to diagnose leukemia as early as possible, increasing the treatment options and patient survivability. Some basic handicraft leukemia detection processes have been introduced in this arena though these are not so accurate and efficient. The proposed approach has been introduced an automated ALL recognition system from the peripheral blood smear. Initially, the color threshold has been applied to segment lymphocytes blood cells from the blood smear. Some post-processing techniques like morphological operation and watershed have been executed to segment the particular lymphocytes cell. Finally, we used a support vector machine (SVM) classifier to classify the cancerous image frames using a statistical feature vector obtained from the segmented image. The proposed framework has achieved the highest accuracy of 99.21%, the sensitivity of 98.45%, specificity of 99%, the precision of 99%, and F1 score of 99.1%, which has beat existing and common states of art methods. We are confident that the proposed approach will positively impact the ALL detection arena.

Lane marking recognition is one of the most crucial features for automotive vehicles as it is the most preliminary requirement of all the autonomy features of the ADAS Systems. The researchers have made a promising improvement in the application of Lane Marking Detection (LMD) in recent times. This research article has taken the initiative to review LMD using deep learning techniques. This paper initially discusses the importance of LMD using Deep Neural Network (DNN) techniques. LMD processes can be categorized into four steps: the dataset, pre-processing, DNN networks, and post-processing. Besides, the network's architecture is described emphasizing particular groups of a neural network like CNN, E-Net, a combination of CNN and RNN, including their contributions and limitations. Besides, comparative performance result with visualization of the final output of five existing techniques is elaborated. Finally, this review is concluded by pointing to some particular challenges future directions for solving the issues.

Nowadays, researchers are incorporating many modern and significant features on advanced driver assistance systems (ADAS). Lane marking detection is one of them, which allows the vehicle to maintain the perspective road lane. Conventionally, it is detected through handcrafted and very specialized features and goes through substantial post-processing, which leads to high computation, and less accuracy. Additionally, this conventional method is vulnerable to environmental conditions, making it an unreliable model. Consequently, this research work presents a deep learning-based model that is suitable for diverse environmental conditions, including multiple lanes, different daytime, different traffic conditions, good and medium weather conditions, and so forth. This approach has been derived from plain encode-decode E-Net architecture and has been trained by using the differential and cross-entropy losses for the backpropagation. The model has been trained and tested using 3,600 training and 2,700 testing images from TuSimple, a robust public dataset. Input images from very diverse environmental conditions have ensured better generalization of the model. This framework has reached a max accuracy of 96.61%, with an F1 score of 96.34%, a precision value of 98.91%, and a recall of 93.89%. Besides, this model has shown very small false positive and false negative values of 3.125% and 1.259%, which bits the performance of most of the existing state of art models.

Image encryption is an important area in visual cryptography that helps in protecting images when shared through internet. There is lot of cryptography algorithms applied for many years in encrypting images. In the recent years, artificial intelligence techniques are combined with cryptography algorithms to support image encryption. Some of the benefits that artificial intelligence techniques can provide are prediction of possible attacks on cryptosystem using machine learning algorithms, generation of cryptographic keys using optimization algorithms, etc. Computational intelligence algorithms are popular in enhancing security for image encryption. The main computational intelligence algorithms used in image encryption are neural network, fuzzy logic and genetic algorithm. In this paper, a review is done on computational intelligence-based image encryption methods that have been proposed in the recent years and the comparison is made on those methods based on their performance on image encryption.

More than eighty-five to ninety percentage of the diabetic patients are affected with diabetic retinopathy (DR) which is an eye disorder that leads to blindness. The computational techniques can support to detect the DR by using the retinal images. However, it is hard to measure the DR with the raw retinal image. This paper proposes an effective method for identification of DR from the retinal images. In this research work, initially the Weiner filter is used for preprocessing the raw retinal image. Then the preprocessed image is segmented using fuzzy c-mean technique. Then from the segmented image, the features are extracted using grey level co-occurrence matrix (GLCM). After extracting the fundus image, the feature selection is performed stochastic gradient descent, and least absolute shrinkage and selection operator (LASSO) for accurate identification during the classification process. Then the inception v3-convolutional neural network (IV3-CNN) model is used in the classification process to classify the image as DR image or non-DR image. By applying the proposed method, the classification performance of IV3-CNN model in identifying DR is studied. Using the proposed method, the DR is identified with the accuracy of about 95%, and the processed retinal image is identified as mild DR.

Blockchain technology is one of the latest innovations that provides a secured decentralized, distributed database of records of all transaction and digital events. In recent years, blockchain technology is being applied in many powerful applications such as financial services, smart contracts, healthcare, supply chain management, e-voting etc. As blockchain ensures high security, it is being trusted and accepted by many business organizations. Consensus algorithms are the primary component of blockchain technology that maintains integrity and security in blockchain. This paper reviews consensus algorithms in blockchain technology that were applied in the recent research works for providing data sharing security. The importance of consensus algorithms in blockchain, their advantages and disadvantages are discussed in this paper.

Monitoring the tool flank wear without affecting the throughput is considered as the prudent method in production technology. The examination has to be done without affecting the machining process. In this paper we proposed a novel work that is used to determine tool flank wear by observing the sound signals emitted during the turning process. The work-piece material we used here is steel and aluminum and the cutting insert was carbide material. Two different cutting speeds were used in this work. The feed rate and the cutting depth were constant whereas the flank wear was a variable. The emitted sound signal of a fresh tool (0 mm flank wear) a slightly worn tool (0.2 -0.25 mm flank wear) and a severely worn tool (0.4mm and above flank wear) during turning process were recorded separately using a high sensitive microphone. Analysis using Singular Value Decomposition was done on these sound signals to extract the feature sound components. Observation of the results showed that an increase in tool flank wear correlates with an increase in the values of SVD features produced out of the sound signals for both the materials. Hence it can be concluded that wear monitoring of tool flank during turning process using SVD features with the Fuzzy C means classification on the emitted sound signal is a potential and relatively simple method., {"references":["E. Dimla, and S. Dimla , \"Sensor signals for tool-wear monitoring in\nmetal cutting operations-a review of methods\", International Journal\nof Machine Tools & Manufacture, vol. 40, pp. 1073- 1098, 2000.","B. Sick, \"On-line and indirect tool wear monitoring in turning with\nartificial neural networks: A review of more than a decade of research\",\nMechanical Systems and Signal Processing, vol. 16, no. 4, pp. 487-\n546, 2002.","J. Kopa, and S. Sali, \"Tool wear monitoring during the turning process\",\nJournal of Materials Processing Technology, vol. 113, Issues 1-3, pp.\n312-316, 15 June 2001.","A. G. Rehorn, J. Jiang, and P. E. Orban, \"State-of the art methods and\nresults in tool conditioning monitoring: a review\", International\nJournal of Advanced Manufacturing Technology , vol. 26, pp. 693-\n710, 2005.","A.Samraj, S. Sayeed, L.C. Kiong., and N. E. Mastorokis, \"Eliminating\nForgers Based on Intra Trial Variability in Online Signature Verification\nUsing Handglove and Photometric Signals\", Journal of Information\nSecurity, vol. 1, pp. 23-28, 2010.","N. S. Kamel, S. Sayeed and G.A. Ellis, \"Glove-based approach to online\nsignature verification\", IEEE Transactions on Pattern Analysis\nand Machine Intelligence, USA, vol. 30, no. 6, pp. 1109-1113, 2008.","S. Sayees, A. Samraj, R. Besar., and L. C. Kiong, \"Forgery Detection\nin Dynamic Signature Verification by Entailing Principal Component\nAnalysis\". Discrete Dynamics in Nature and Society, vol. 2007 , pp. 1-\n8, 2007.","SVD and signal processing: Algorithms, Applications and\nArchitectures,(1989) F. Deprettere, ed., North Holland Publishing Co.","E.Biglieri, and K. Yao, \"Some properties of singular value\ndecomposition and their applications to digital signal processing\",\nSignal Processing, vol. 18, Issue 3, pp. 277-289, November 1989,.\n[10] J. Bezdek, \"Pattern Recognition with Fuzzy Objective Functions\",\nPlenum, New York, 1981.\n[11] Mohanad A., Mohammad M., Abdullah R., \"Optimizing of Fuzzy CMeans\nClustering Algorithm Using GA\", World Academy of Science,\nEngineering and Technology, Vol. 39, 2008.\n[12] Bainian Li, Kongsheng Zhang, and Jian Xu, \"Similarity measures and\nweighed fuzzy c-mean clustering algorithm\", International Journal of\nElectrical and Computer Engineering, vol. 6, no. 1, pp. 1-4. 2010."]}

The Neuro-Fuzzy hybridization scheme has become of research interest in pattern classification over the past decade. The present paper proposes a novel Modified Adaptive Fuzzy Inference Engine (MAFIE) for pattern classification. A modified Apriori algorithm technique is utilized to reduce a minimal set of decision rules based on input output data sets. A TSK type fuzzy inference system is constructed by the automatic generation of membership functions and rules by the fuzzy c-means clustering and Apriori algorithm technique, respectively. The generated adaptive fuzzy inference engine is adjusted by the least-squares fit and a conjugate gradient descent algorithm towards better performance with a minimal set of rules. The proposed MAFIE is able to reduce the number of rules which increases exponentially when more input variables are involved. The performance of the proposed MAFIE is compared with other existing applications of pattern classification schemes using Fisher-s Iris and Wisconsin breast cancer data sets and shown to be very competitive., {"references":["P.K. Simpson, \"Fuzzy Min-Max Neural Networks-Part\n1:Classification,\" IEEE Transaction on Neural Networks, vol. 3, no.5,\npp.776-786, Sept. 1992.","S. Abe and M.S. Lan, \"Fuzzy Rules Extraction Directly from Numerical\nData for Function Approximation,\" IEEE Transaction on System, Man,\nand Cybernetics, vol. 25, no.1, pp.119-129, Jan. 1995.","G.O.A. Zapata, R.K.H. Galvao, and T. Yoneyama, \"Extracting Fuzzy\nControl Rules from Experimental Human Operator Data,\" IEEE\nTransaction on System, Man and Cybernetics - Part B: Cybernetics, vo.\n29, no. 3, pp 25-40, Feb. 1999.","Han, J. and Kamber, M., Data Mining: Concepts and Techniques,\nSecond Edition, Morgan Kaufmann publishers, San Francisco, 2006","R. Agrawal, R. Srikant, \"Fast Algorithms for Mining Association\nRules\", Proceedings of the 20th VLDB Conference, Santiago, Chile,\n1994.","Bilal I. S., Keshav P. D., Alamgir M. H., Mohammad S A., \"\nDiversification of Fuzzy Association Rules to Improve Prediction\nAccuracy\", Fuzzy Systems (Fuzz) in IEEE Explorer, 2010.","T. Takagi and M. Sugeno, \"Fuzzy identification of systems and its\napplications to modeling and control,\" IEEE Transaction on Systems,\nMan, and Cybernetics, vol. SMC-15, pp. 116-132, Jan.-Feb. 1985","X. Zeng and M. G. Singh,, \"Approximation Theory of Fuzzy Systems-\nMIMO Case IEEE Transactions on Fuzzy Systems, vol. 3, no. 2, May\n1995.","James C. Bezdek, Pattern Recognition with Fuzzy Objective Function\nAlgorithms, Plenum Press, pp.65-86, 1981.\n[10] Mohanad A., Mohammad M., Abdullah R., \" Optimizing of Fuzzy CMeans\nClustering Algorithm Using GA\", World Academy of Science,\nEngineering and Technology, Vol. 39, 2008.\n[11] C.L. Blake and C.J. Merz, \"UCI Repository of Machine Learning\nDatabases,\" University of California, Irvine, Department of Information\nand Computer Science, http://www.ics.uci.edu/~mlearn\n/MLRepository.html, 1998\n[12] S. Fahlman and C. Lebiere, \"The Cascade-Correlation Learning\nArchitecture,\" Carnegie Melloin Univ., School of Computer Science,\nTechnical Report CMU-CS- 90-100, Feb. 1990.\n[13] T-P. Hong and S.-S. Tseng, \"A Generalised Version Space Learning\nAlgorithm for Noisy and Uncertain Data,\" IEEE Transaction on\nKnowledge and Data Eng., vol. 9, no. 2, pp. 336-340, Mar.-Apr. 1997.\n[14] S.C. Newton, S. Pemmaraju, and S. Mitra, \"Adaptive Fuzzy Leader\nClustering of Complex Data Sets in Pattern Recognition,\" IEEE\nTransaction on Neural Networks, vol. 3, no.5, pp.794-800, Sept. 1992.\n[15] T.P. Wu and S.M. Chen, \"A New Method for Constructing Membership\nFunctions and Fuzzy Rules from Training Examples,\" IEEE Transaction\non System, Man, and Cybernetics - Part B: Cybernetics, vol. 29, no.1,\npp.25-40, Feb. 1999.\n[16] R. Setiono, \"Extracting M-of-N Rules from Trained Neural Networks,\"\nIEEE Transaction On Neural Networks, vol. 11, no. 2, pp.512- 519, Mar.\n2000.\n[17] B.C. Lovel and A.P. Bradley, \"The Multiscale Classifier,\" IEEE\nTransaction On Pattern Analysis and Machine Intelligence, vol. 18, no.\n2, pp. 124-137, Feb. 1996.\n[18] H.-M. Lee, C.-M. Chen, J.-M. Chen, and Y.-L. Jou, \"An Efficient Fuzzy\nClassifier with Feature Selection Based on Fuzzy Entropy,\" IEEE\nTransaction on Systems, Man, and Cybernetics - Part B: Cybernetics,\nvol. 31, no.3, pp.426-432, June 2001.\n[19] A. Chatterjee and A. Rakshit, \"Influential Rule Search Scheme (IRSS) -\nA New Fuzzy Pattern Classifier,\" IEEE Transaction on Knowledge and\nData Engineering, vol. 16, no. 8, pp. 881-893Aug. 2004.\n[20] ChangSu L., Anthony. Z., Tomas B., \"An Adaptive T-Stype Rough-\nFuzzy Inference System (ARFIS) for Pattern Classification\", Fuzzy\nInformation Society, IEEE Explorer, pp. 117-122, 2007.\n[21] Sandeep C., and Rene V. M., \"RANFIS: Rough Adaptive Neuro-Fuzzy\nInference System\", International Journal of Computational Intelligence,\nvol. 3, No. 4, 2006."]}