Browsing by Subject "Hidden Markov Models"

Now showing 1 - 5 of 5

Open Access
Finite perturbation analysis methods for optimization of inventory systems with non-stationary Markov-modulated demand and partial information
(2018-01) Güleçyüz, Süheyl
The state of the economy may fluctuate due to several factors, and the customer demand is a affected from the fluctuations of the state of the economy. Although the inventory holders can predict the state of the economy based on the demand realizations, they generally do not have the true state information. The lack of information can be extended to the transition probabilities in the state, and the demand distributions associated with each state. Further extensions may include the actual number of demand states. We consider a single-item, periodic-review inventory system with Markov-modulated discrete-valued demand, constant lead time, and full backlogging. The true demand distribution state is partially observed based on the realized demands. We study the infinite horizon average cost minimization problem, in which the optimal inventory replenishment policy is a state-dependent base-stock policy. We develop a local search method based on finite perturbation analysis (FPA) to find the base-stock levels for a finite number of discretized state beliefs. We then extend our search method to the unknown transition matrix and demand distribution case. We compare the FPA-based local search algorithm with a myopic base-stock policy, the Viterbi algorithm, and the sufficient statistics method, in terms of the average cost. Finally, we analyze how the average cost changes with respect to the estimated number of demand states when the actual number of states is unknown.
Open Access
Fire detection algorithms using multimodal signal and image analysis
(2009) Töreyin, Behçet Uğur
Dynamic textures are common in natural scenes. Examples of dynamic textures in video include fire, smoke, clouds, volatile organic compound (VOC) plumes in infra-red (IR) videos, trees in the wind, sea and ocean waves, etc. Researchers extensively studied 2-D textures and related problems in the fields of image processing and computer vision. On the other hand, there is very little research on dynamic texture detection in video. In this dissertation, signal and image processing methods developed for detection of a specific set of dynamic textures are presented. Signal and image processing methods are developed for the detection of flames and smoke in open and large spaces with a range of up to 30m to the camera in visible-range (IR) video. Smoke is semi-transparent at the early stages of fire. Edges present in image frames with smoke start loosing their sharpness and this leads to an energy decrease in the high-band frequency content of the image. Local extrema in the wavelet domain correspond to the edges in an image. The decrease in the energy content of these edges is an important indicator of smoke in the viewing range of the camera. Image regions containing flames appear as fire-colored (bright) moving regions in (IR) video. In addition to motion and color (brightness) clues, the flame flicker process is also detected by using a Hidden Markov Model (HMM) describing the temporal behavior. Image frames are also analyzed spatially. Boundaries of flames are represented in wavelet domain. High frequency nature of the boundaries of fire regions is also used as a clue to model the flame flicker. Temporal and spatial clues extracted from the video are combined to reach a final decision.Signal processing techniques for the detection of flames with pyroelectric (passive) infrared (PIR) sensors are also developed. The flame flicker process of an uncontrolled fire and ordinary activity of human beings and other objects are modeled using a set of Markov models, which are trained using the wavelet transform of the PIR sensor signal. Whenever there is an activity within the viewing range of the PIR sensor, the sensor signal is analyzed in the wavelet domain and the wavelet signals are fed to a set of Markov models. A fire or no fire decision is made according to the Markov model producing the highest probability. Smoke at far distances (> 100m to the camera) exhibits different temporal and spatial characteristics than nearby smoke and fire. This demands specific methods explicitly developed for smoke detection at far distances rather than using nearby smoke detection methods. An algorithm for vision-based detection of smoke due to wild fires is developed. The main detection algorithm is composed of four sub-algorithms detecting (i) slow moving objects, (ii) smoke-colored regions, (iii) rising regions, and (iv) shadows. Each sub-algorithm yields its own decision as a zero-mean real number, representing the confidence level of that particular subalgorithm. Confidence values are linearly combined for the final decision. Another contribution of this thesis is the proposal of a framework for active fusion of sub-algorithm decisions. Most computer vision based detection algorithms consist of several sub-algorithms whose individual decisions are integrated to reach a final decision. The proposed adaptive fusion method is based on the least-mean-square (LMS) algorithm. The weights corresponding to individual sub-algorithms are updated on-line using the adaptive method in the training (learning) stage. The error function of the adaptive training process is defined as the difference between the weighted sum of decision values and the decision of an oracle who may be the user of the detector. The proposed decision fusion method is used in wildfire detection.
Open Access
Ionogram scaling using Hidden Markov Models
(Institute of Electrical and Electronics Engineers, 2018) Gök, Gökhan; Alp, Y. K.; Arıkan, Orhan; Arikan, F.
In this paper, a novel method for electron density reconstruction using ionosonde data is proposed. Proposed technique uses Hidden Markov Models for extracting echoes that provides valuable information about electron density distribution in order to provide input to a model based optimization technique that reconstructs the electron density distribution by solving model parameters. Analysis on real ionosonde data shows that proposed technique outperforms standard techniques in the literature.
Open Access
A method for automatic scaling of ionograms and electron density reconstruction
(IEEE, 2021-10-19) Gök, Gökhan; Alp, Y. K.; Arıkan, Orhan; Arıkan, F.
Ionogram scaling is the process of reconstructing electron density with respect to height by using the measurements of a remote sensing instrument known as ionosonde. In this study, a novel two stage ionogram scaling technique, ISED, is proposed. In the first stage, Hidden Markov Models (HMMs) are used to identify the actual ionospheric reflections in the ionosonde measurements. In the second stage, an IRI-Plas model based optimization problem is solved to obtain the vertical profile that generates the best least squares fit to the reflections identified in the first stage. To show the performance of ISED in global scale, experiments are conducted on 14,812 ionograms recorded at the three different stations which are Pruhonice in Czech Republic, Eielson in USA and Sao Luis in Brazil. Application of ISED to raw ionograms indicate 97.6% of the cases, ISED provides accurate electron density reconstructions, which is an improvement about 8.7% over ARTIST, most commonly used ionogram scaling technique.
Open Access
Pyroelectric infrared (PIR) sensor based event detection
(2009) Soyer, Emin Birey
Pyroelectric Infra-red (PIR) sensors have been extensively used in indoor and outdoor applications as they are low cost, easy to use and widely available. PIR sensors respond to IR radiating objects moving in its viewing range. The current sensors give an output of logical one when they detect a hot object’s motion and a logical zero when there is no moving hot object. In this method, only moving objects can be detected and the rate of false alarm is high. New types of PIR sensors are more sophisticated and more capable. They have a lower false alarm ratio compared to classical ones. Although they can distinguish pets and humans, again they can only be used for detection of hot object motions due to the limitations caused by the usage of the simple comparator structure inside. This structure is unalterable, not flexible for development, and not suitable for implementing algorithms. A new approach is developed to use PIR sensors by modifying the sensor circuitry. Instead of directly using the output of a classical PIR sensor, an analog signal is extracted from the PIR output and it is sampled. As a result, intelligent signal processing algorithms can be developed using the discrete-time sensor signal. In this way, it is possible to develop human, pet and flame detection methods. It is also possible to find the direction of moving objects and estimate their distances from the sensor. Furthermore, the path of a moving target can be estimated using a PIR sensor array. We focus on object and event classification using sampled PIR sensor signals. Pet, human and flame detection methods are comparatively investigated. Different human motion events are modeled and classifed using Hidden Markov Models (HMM) and Conditional Gaussian Mixture Models (CGMMs). The sampled data is wavelet transformed for feature extraction and then fed into HMMs for analysis. The final decision is reached according to the Markov Model producing the highest probability. Experimental results demonstrate the reliability of the proposed HMM based decision and event classification algorithm.