Informatics Publishing Limited

Tarun Dhar Diwan ¹, Priti Verma ², Kamal Mehta ³

Affiliations
1 Deptt. of Engineering, Dr. C. V. Raman University, Bilaspur (C.G), IN
2 Dr. C. V. Raman University, Bilaspur (C.G), IN
3 Sagar Institute of Sciences and Technology, Bhopal, IN

Abstract

Sentence boundary detection system has three independent applications (Rule-based, HMM, and Maximum Entropy). Maximum Entropy Model is the central part of this system, which achieved an error rate less than 2% on part of the Wall Street Journal (WSJ) Corpus with only eight binary features. The performance of the three applications is illustrated and discussed. Sentence boundary disambiguation is the task of identifying the sentence elements within a paragraph or an article. Because the sentence is the basic textual unit immediately above the word and phrase, Sentence Boundary Disambiguation (SBD) is one of the essential problems for many applications of Natural Language Processing – Parsing, Information Extraction, Machine Translation, and Document Summarizations. The accuracy of the SBD system will directly affect the performance of these applications. However, the past research work in this field has already achieved very high performance, and it is not very active now. The problem seems too simple to attract the attention of the researchers.

Keywords

Sentence Boundary Disambiguation, Maximum Entropy Model, Features, Generalized Iterative Scaling, Hidden Markov Model.

Full Text

     

Tarun Dhar Diwan ¹, Sumati Gauraha ², Kamal Mehta ³

Affiliations
1 Deptt. of Engineering, Dr. C. V. Raman University, Bilaspur (C.G), IN
2 Dr. C. V. Raman University, Bilaspur (C.G), IN
3 Sagar Institute of Sciences and Technology, Bhopal, IN

Abstract

Driver in-alertness is an important cause for most accidents related to vehicle crashes. Drowsy driver detection methods can form the basis of a system to potentially reduce accidents related to driver doziness. Intel-Eye describes a real-time driver in alertness and shock related facial expression monitoring. Intel -Eye obtains visual cues such as eyelid movement; gaze movement, head movement, and facial expression that typically characterize the level of alertness of a person are extracted in real time and systematically combined to infer the fatigue level of the driver. Intel-eye distinguishes itself by the Two-Way Approach in eye gaze analysis. Shock analysis is done to identify the driver’s expression and signals are sent for automatic braking system. A probabilistic model is developed to model in Intel-Eye and it is used for predicting human in-alertness based on the visual cues obtained. But this model is mainly focus on eye blinking because this system is connected with programming by using METLAB. I have used MATLAB programming to make an automatic eye blink tracking and detection system for a video. The eyes are tracking by image sequence, the eyes are tracked and correlation scores between the actual eye and the corresponding “closed-eye” templates, which are used to detect blinks. Accurate head and eye tracking results are obtained at a processing rate of more than 30 frames per second (fps), in more than 90% cases with a low false positive blink detection of 0.01%. I take ideal eye blinking rate in first 10 minutes in driving of human, then we observe the changes rate in blinking frequencies. We observed that the dangerous condition occurs when the eye blinking rate are decreases (increase) as 50% (50%), 75% (100%), 100% (300%) from natural condition for lower , medium and higher level dangerous conditions respectively.

Keywords

Measurement of Eye Blinking, Sensing of physiological characteristics, Advanced Emergency Braking Systems (AEBS), Electronic Stability Control.

Full Text

     

Kamal Mehta ¹

Affiliations
1 Department of Zoology, J.C.D.A.V College, Dasuya – 144205, Punjab, IN

Abstract

Objective: Toxicity of contaminants in the aquatic ecosystems is found to be affected by rise in temperature due to global warming. The main objective of the present study was to investigate the toxicity effects of some specific organic and inorganic contaminants like iron, copper, endosulfan, chlorpyrifos, phenol, chlorine, cyanides, nitrites and ammonia at different temperatures to fish fauna. Methodology: secondary data concerning with the impact of temperature fluctuations on toxicology of contaminants in aquatic ecosystems especially with reference to fish fauna was collected to review and predict their correlation. Findings: This review study indicated that toxicity of contaminants is greatly affected by climate change and it may have adverse effects on biology of fish fauna. The concentration of toxic contaminants which is not active at low and moderate temperature proved lethal with increase in water temperature. Application: This review study may sensitize the ecologists to draw correlation between temperature and toxicity of contaminants for effective management of aquatic ecosystems and fish fauna. This will also essentially helpful in understanding of basic mechanisms that allow the aquatic fauna to survive in the contaminated waters. Conclusion: The study of interaction mechanisms of aquatic toxicants with environmental factors and their management in aquatic ecosystems, hence, is the need of hour otherwise; they can lead to loss of aquatic biodiversity and can disturb ecological balance.

Keywords

Aquatic Fauna, Contaminants Toxicity, Heavy Metals, Pesticides, Temperature.

Full Text