Engineer Quality

DR. HOOSHMAND M. KALAYEH

** ******** *****, *********, ** 14534 • 585-***-****

***********@*****.***

OBJECTIVE

To contribute in research, development, teaching, and applications of image processing, image analysis, image quality, statistical pattern recognition, neural networks, computer vision, passive and active remote sensing, and medical imaging.

MAJOR ACCOMPLISHMENTS

• Initiated the application of a neural networks program for the design of novel biological active chemicals.

• Implemented a Back Propagation Neural-network software program based on Conjugate Gradient and Stiff methods to enhance the process of chemical compound discovery. This work subsequently led to discovery of a new compound.

• Developed several machine vision systems for process and product understanding, monitoring, and improvement

• Developed a sand grain sample mounting method

• Developed algorithm using shape information to classify agricultural articles

• As an strategic intellectual property coordinator, evaluator, and generator I have filed several patents (please see my patents)

• Developed and implemented several digital image noise reduction, enhancement, and reconstruction algorithms in x-ray radiography, fluoroscopy, tomosynthesis, dual energy, and exposure index

• Developed several Adaptive Connected Component Labeling Algorithms using MATLAB

EXPERIENCE (in brief)

L3-Communications (Security Detection Systems) St. Petersburg, FL 2009 - 2010

Principle Engineer

Infimed, Liverpool, NY January 2008- July 2009

Senior Medical Image Specialist

ITT, Rochester, NY 2000-August 2006

Senior Imaging Scientist

Image Computing Consulting, Rochester, NY August 2006-January 2008

President & CEO

SRC-Vision, Medford, OR 1998-2000

Senior Staff Scientist

E. I. DuPont, Wilmington, DE 1984-1998

Senior Research Engineer

EDUCATION

Purdue University

Doctor of Philosophy, Electrical Engineering, - G.P.A. 5.9/6.0

Wayne State University

Masters of Science, Electrical and Computer Engineering - G.P.A 3.9/4.0

ISSUED PATENTS

1. United States Patent 6822742. System and method for remote quantitative detection of fluid leaks from a natural gas or oil pipeline. Issued 11/23/04

2. United States Patent 6410872. Agricultural article inspection apparatus and method employing spectral manipulation to enhance detection contrast ratio. Issued 6/25/02

3. United States Patent 5579409. Methods for determining the exterior points of an object in a background. Issued 11/26/96

4. Multi-line tunable laser system

5. Ground surface cover type reflectivity variation correction in a differential absorption LIDAR system

6. Synthetic Aperture Radar image compression

7. Multi-Sensor and Differential Absorption LIDAR data fusion

8. Method for improving the performance accuracy in Differential Absorption LIDAR for oil and gas pipeline leak detection

PENDING PATENTS

1. Image stitching in digital radiography

2. Auto-shutter/auto- Region Of Interest Detection in digital radiography

SELECTED PUBLICATIONS (OUT OF MORE THAN 40 PAPERS & TECHNICAL REPORTS)

H. M. Kalayeh, “Electronic I, II and III Books”, Iran University of Science and Technology, 1997

H. M. Kalayeh, D. A. Landgrebe, “Adaptive Relaxation Labeling”, IEEE Transactions on PAMI, Vol., 6, No.3, 1984

H. M. Kalayeh, D. A. Landgrebe “A Stochastic Model Utilizing Spectral and Temporal Characteristics”, IEEE Transactions on GRS, Vol. 9, GE-24, No. 5 1986

H. M Kalayeh, D. A. Landgrebe, “A Stochastic Model Utilizing Spectral and Spatial Characteristics”, IEEE Transaction on PAMI, Vol. 9 No. 3 1987

T. A. Andrea, H. M. Kalayeh, “Analysis of Quantitative Structure Activity Relationships of Dihydrofolate Reductase Inhibitors”, Journal of Medicinal Chemistry, September 1991

H. M. Kalayeh, “Sand Ores Mineral Image Classifier System”, DuPont Technical Report, 1998

T. A. Andrea, H. M. Kalayeh, “Natural products as insecticides: the biology, biochemistry and quantitative structure–activity relationships of spinosyns and spinosoids, John Wiley & Son, Ltd, October 2001

H. M. Kalayeh, “Whole Order Image Orientation Algorithm to Accurately Classify the Entire Customer Orders”, Kodak Technical Report, 2002

SUMMARY OF QUALIFICATIONS

Computer vision systems

• Image acquisition (lighting, optics, CCD color camera)

• Image analysis (Image processing, segmentation, mathematical morphology, and local object feature extraction)

Neural Network

• Statistical pattern recognition

• Non-linear classification and modeling

• Back propagation neural networks

Digital medical Image Processing

• Noise Reduction and Image Enhancement Algorithms in digital x-ray Fluoroscopy

• Noise Reduction and Image Enhancement Algorithms in digital x-ray radiography

• Flat Panel Detector characterization (Defect Detection & Correction, Offset and Gain Correction)

• Quantitative Characterization of Image Quality

• Image Stitching

• Auto-Shutter/Auto- Region Of Interest Detection

• Tomosynthesis

• Dual Energy

• Exposure Index

Airline Luggage Explosive Detection Systems

• CT Based Image Acquisition,

• Image Preprocessing

• Image Segmentation

• Image Feature Extraction

• Image and Object Classification

Remote Sensing

• Hyperspectral image pattern recognition

• Differential Absorption LIDAR

Software

• Prototyping using MATLAB

• Software programming (C/C++)

• Knowledge of WIN NT, MAC-OS, and UNIX

Leadership

• Mentoring junior scientists and collaborating with leaders in promoting a congenial work place community

EXPERIENCE (more detail)

L3-Communications (Security Detection Systems) St. Petersburg, FL 2009 - 2010

Principle Engineer

• Developed several Adaptive Connected Component Labeling Algorithms using MATLAB

• Evaluated the performance of the developed algorithms using 12000 Bags

• Reduced the False Alarm without reducing the minimum required Potential Threat per class accuracy

EXPERIENCE (more detail)

Infimed, Liverpool, NY January 2008- July 2009

Senior Medical Image Specialist

• Developed and implemented several digital image noise reduction, enhancement, and reconstruction algorithms in x-ray radiography, fluoroscopy, tomosynthesis, dual energy, and exposure index

• Noise Reduction and Image Enhancement Algorithms in digital x-ray Fluoroscopy

• Noise Reduction and Image Enhancement Algorithms in digital x-ray radiography

• Flat Panel Detector characterization (Defect Detection & Correction, Offset and Gain Correction)

• Quantitative Characterization of Image Quality

• Image Stitching

• Auto-Shutter/Auto- Region Of Interest Detection

• Tomosynthesis (Using Shift And Add method for reconstruction)

• Dual Energy (Segment tissues from bones)

• Exposure Index (Determine the minimum X-ray does for an acceptable image quality)

Image Computing Consulting, Rochester, NY August 2006-January 2008

President & CEO

• Started my own Online Image Computing Consulting

ITT, Rochester, NY 2000-August 2006

Senior Imaging Scientist

• Intellectual property coordinator, evaluator, and generator (please see my patents)

• Expert resource in multispectral image processing and pattern recognition algorithms.

• See my related issued and pending patents.

• Developed and implemented a modular whole order picture orientation algorithm to accurately classify the entire corporate customer order system using C/C++. Reduced original 10,000 line of code to 500 lines and the same time made it modifiable, readable, expandable, and modular. This algorithm was first prototyped using MATLAB before the C/C++ implementation.

• Temperature and Pressure Dependence of the Cross Section: Effect on the Error.

• 3-Line laser DIAL Concentration Path Length Methodologies simulated using MATLAB

• Concentration Path Length (CPL) Variance and SNR Due to the Additive Noise.

• Independent Surface Reflectivity Plume Detection Methodology.

• Classification of Invisible Stressed Trees Using Hyperspectral Image Data.

DR. HOOSHMAND M. KALAYEH 585-***-****

SRC-Vision, Medford, OR 1998-2000

Senior Staff Scientist

Senior technical advisor to the president of the company

• Developed algorithm using shape information to filter, sort, and classify agricultural articles (onion rigs, french fried, raisins and chicken pieces) based on line scan camera and X-ray images. Demonstrated performance accuracy of the system using KHOROS and worked with software team to transfer the algorithm for C/C++ implementation.

• Developed algorithm using shape information to filter out the touching object and prototype the algorithm using KHOROS for a statistical monitoring of french-fries based on the acquired images of a line scan camera. I demonstrated the performance accuracy of the system and then worked with the software team for the transfer of the algorithm.

• Developed algorithm using shape information to classify onion rings and then prototyped the algorithm using KHOROS based on the acquired images of a line scan camera. I demonstrated the performance accuracy of the system and then worked with the software team for the transfer of the algorithm for a real time implementation.

• Developed algorithm using shape information to sort different chicken pieces based on an X-ray acquired images. I prototype the algorithm using KHOROS. I demonstrated the performance accuracy of the system and then worked with the software team for the transfer of the algorithm for a real time implementation.

EXPERIENCE (more detail Continued)

E. I. DuPont, Wilmington, DE 1984-1998

Senior Research Engineer

• Developed several machine vision systems for process and product understanding, monitoring, and improvement

• Initiated the application of a neural networks program for the design of novel biological active chemicals.

• Implemented a Back Propagation Neural-network (BPN) software program (FORTRAN) based on Conjugate Gradient and Stiff methods to enhance the process of chemical compound discovery. This work subsequently led to discovery of a new compound.

• Developed automatic pellet image analyzer to estimate weight distributions, eliminating manual time consuming product quality monitoring and enabling an operator to efficiently check more products. The key aspect of this developed algorithm was based on image segmentation and different grain color features extraction and classification of each grain. The algorithm is implemented in C.

• Developed a particle mounting method using image segmentation and mathematical morphologies to separate the touching particles which enabled subsequent classification and size distributions.

• As part of a team, developed PC-UNIX based color sand ores image analyzer to gain knowledge of heavy mineral grade distribution in ground and separation processes which improved efficiency. The objective of this program was to mount nine colored grains, acquire images of thousands of grains, segment each grain from the background, classify each grain based on its shape and color features, count the number of each grain type and estimate it weigh distribution. The software aspect of the program developed based on KHOROS and C.

• Developed missing or gained carbon black filament in P-140 image analyzer to eliminate unwanted production. A missing and gained carbon back filament was adaptively classified based on its circular local texture.

• Developed a PC based laser far-field image pattern classifier to select the right laser for an optical modulator. The elliptical pattern (shape) of the far-field patent extracted from the noise and its corresponding parameters were estimated for the selection of a laser.

• Developed a plant disease image analyzer to improve consistently and accurately measure the efficacy of applied compounds. There were fourteen different plant diseases. Each plant was a 3D object with many leaves and different diseases were affected the leaves. Images of each plant were acquired by a color camera as the plant was rotated. I used mathematical morphologies to separate the target (local diseased regions on each leaf) from the background, classify each region, and then estimate the total area of the plant affected by a disease.

• As part of a team, developed laser scanner based Tyvek® uniformity analyzer to improve product quality. In this system local textural features are extracted and used to classify the uniformity of a roll of Tyvek.

• Developed an image analysis system to improve an on-line extruder mixing and melting process. This system predicted extruded product quality and diagnosed extruder-related problems based on uniformity of the mixing. The uniformity of the mixing was characterized based on the acquired image texture.

• As part of a team, developed a speedometer inspection system to monitor its correct reading based on image segmentation and registration.

• Developed a Lyrca® image acquisition system classifying its Grin/Glitters used in quality control to prevent the production of unwanted products. Two images were acquired by two different light sources, subtracted, and its texture was extracted to solve this problem.

• Developed a KHOROS based touching rock ores image segmentation to classify rock ores and estimate their size distributions.

• Developed a KHOROS based gangue minerals image classifier to predict the state of a chlorinator and prevent costly operation shutdown.

EXPERIENCE (more detail Continued)

Science and Technology University

Electronic Engineering: Graduated with Honors.

• Taught electronic courses to under graduates.

• Wrote Electronic I, II and III Books.

• Developed Electronic I, II, and III laboratories.

Contact this candidate