Hong Zhao
Objectives
Address:
Canton, MI, 48187
Email: acuy72@r.postjobfree.com
Cell phone: 734-***-****
LinkedIn: https://www.linkedin.com/in/hong-zhao-539b7164 Seeking a research/engineering position in the related areas of Audio/Voice Signal Processing and Recognition, Image/Video Processing, Machine Learning.
Education
Sept. 2007 to Ph.D. (Obtained Ph.D. degree on Apr. 2013) Apr. 2013 Southwest Jiaotong University
• Major: Information and Communication Engineering
• Research Areas: Multimedia Security, Image Processing, Audio Forensics, Signal Processing Oct. 2010 to Visiting Ph.D. student
Sept. 2012 University of Michigan - Dearborn
• Research Areas: Multimedia Security, Audio Forensics, Signal Processing Sept. 2003 to Undergraduate Student
Jun. 2007 Southwest Jiaotong University
• Major: Information Security
Work Experience
Jun. 2015 to Postdoctoral Research Fellow
Mar. 2016 University of Michigan-Dearborn
Leading and working within a close-knit team on intelligent system and algorithm design, including driver fatigue detection using Matlab and LibSVM, object classification/recognition using Matlab, LibSVM, and VLFeat toolbox, lane change detection using Matlab, LibSVM, and Neural Network. Also providing technical and academic assistance to the undergraduate and graduate students. May 2013 to Postdoctoral Research Fellow
Dec. 2014 South University of Science and Technology of China Working on the topic of physical layer security, wireless channel estimation, transmitter/receiver identification. Designing a device authentication algorithm by using the device fingerprint and the response of the transmission channel. All programs were implemented in Matlab.
Project Experience and Key Skills
Signal Processing
Experience
Using FFMEPG to decode and encode video and audio, e.g., H.264, and MP3 files.
Using Libjpeg to decode and encode JPEG image.
Familiar with image/video/audio processing algorithms
Familiar with popular filtering algorithms, such as Kalman filtering, particle filtering
Familiar with signal processing toolbox in Matlab, audio signal processing toolbox (voicebox) Audio Signal
Processing
Audio Forensics project: I am responsible to design an audio classification system, which can distinguish authentic audio and forged audio. Statistical modeling and feature extraction methods such as short time Fourier transform, MFCC modulation frequency domain-based features are used. Machine learning method such as SVM and GMM is applied for classification. During this project, I earned audio signal processing experience, such as the audio compression standard, audio feature extraction methods, psychoacoustics model of human hearing system, audio noise cancellation algorithms, dereverberation algorithms and so on.
Statistical
Learning
Experience
Fluent in using statistical learning algorithms (e.g., Support vector machine, neural network, random forest, ensemble classifier) for pattern recognition, classification, and prediction
Fluent in using statistical probability models (e.g., Gaussian mixed model, linear/non-linear regression) for data ming, model discovery, and inference Project Experience and Key Skills- continued
Driver/Vehicle
Signal Analysis
Driver fatigue detection project (2015-2016): I am responsible to design a driver fatigue detection system by using machine learning methods (in Maltab). Physiological signals are recorded by the recording sensors. Statistical analysis methods are applied on the recorded signals to extract the features. SVM is used to learn the system parameters and to evaluate the system performance.
Lane departure detection project (2015-2016): This project uses vehicle signals (such as steering angle) and front view video for lane departure detection. Statistical analysis and feature extraction methods are implemented in Matlab. Also, SVM is used to train the system and to detect the lane departure behavior.
Involved in V2V communication, collision warning system based on V2V Computer Vision
Experience
I was involved in image classification and image authentication projects from 2007 to 2011. I am responsible to design new image classification and image authentication methods using Matlab and Libsvm. Experience gained from this project includes image pre/post-processing, statistical feature extraction, and classification system design. These experience are useful for other application, such as facial recognition, image restoration, image denoising, etc.
Knowledge of deep learning algorithms (Convolutional Neural Networks, Matconvnet) for object recognition, image classification
Fund
Experience
Leader of two national funds (experience in proposal/final report writing, algorithm design and evaluation) (2010-2012, 2015-2017)
Leader of driver wellness and lane departure detection projects (experience in supervising the fresh team members) (2015-2016)
Data Security
Experience
Knowledge of popular cryptography, encryption/decryption algorithms (IDEA, RSA, DES, etc)
Knowledge of data security, privacy protection, data integrity authentication Programming
Languages
Matlab programming (2007 - now), Python (2012-now), C/C++ (2007-2012) Selected Publications
• Hong Zhao, Yifan Chen, Rui Wang, Hafiz Malik, Anti-Forensics of Environmental- Signature-Based Audio Splicing Detection and Its Countermeasure via Rich-features Classification, IEEE Transactions on Information Forensics and Security, March, 2016, Accepted
• Keren Wang, Hong Zhao, Hongxia Wang, Video Steganalysis against Motion Vector Based Steganography by Adding or Subtracting One Motion Vector Value, IEEE Trans- actions on Information Forensics and Security, 2014, 9(5), 741-751
• Hong Zhao, Hafiz Malik, Audio Recording Location Identification Using Acoustic Environment Signature, IEEE Transactions on Information Forensics and Security, 2013, 8(11), 1746–1759
• Jiaqi Ma, Yilu Murphey, Hong Zhao, Real Time Drowsiness Detection Based on Lateral Distance Using Wavelet Transform and Neural Network, IEEE Symposium Series on Computational Intelligence, 2015, 411 – 418
• Hong Zhao, Yifan Chen, Rui Wang, Hafiz Malik, Audio Source Authentication and Splicing Detection using Acoustic Environmental Signature, the 2nd ACM Workshop on Information Hiding and Multimedia Security(IH&MMSec), 2014, 159– 164
• Hong Zhao, Hafiz Malik, Audio Forensics using Acoustic Environment Traces, IEEE Statistical Signal Processing Workshop
(SSP), 2012, 373–376
Referees
Dr. Yilu Murhpey Dr. Hafiz Malik Dr. Yifan Chen
Professor and Associate Dean Associate Professor Professor and Department Chair Department of Electrical and
Computer Engineering
Department of Electrical and
Computer Engineering
Department of Electrical and
Electronic Engineering
University of Michigan-Dearborn University of Michigan-Dearborn South University of Science and Technology of China Dearborn, MI, 48128 Dearborn, MI, 48128 Shenzhen, China, 518055
Tel: (1-313-***-**** Tel: (1-313-***-**** Tel: +86-075*-******** Email: acuy72@r.postjobfree.com Email: acuy72@r.postjobfree.com Email: acuy72@r.postjobfree.com