TWO DAYS WORKSHOP ON DEEP LEARNING AND IMAGE PROCESSING FOR MEDICAL AND INDUSTRIAL CONTROL APPLICATIONS.

---

---

Inaugural Address was given by  Principal Dr. N. S. Bhuvaneswari. The program started with first Guest speaker, Dr.B. Hema Kumar, Assistant Professor, Pondicherry Engineering College delivered his lecture on MEDICAL IMAGE PROCESSING WITH PIXELS covering topics -

• Improvement of pictorial information for human interpretation.
• Signals are one dimensional while images are in 2 dimensional.
• Color images consist of RGB each Plane has 0 to 255 range (2^8) 8 bit representation.
• Images appear in different color than the actual which actually depends on surrounding colors impact : Tonal Resolution.
• MATLAB operations on array data type data’s.
• Increment in rows, extracting first row, first column, row, magic matrix, reading image.
• When RGB has same Value of intensity it becomes gray image.
• Image histogram is used for gray images.
• When intensities are widely distributed implies good contrast of image. Contrast can be improved by equalization, piecewise distribution.
• Masking process. Salt and pepper noise compensated by median filter.

HANDS ON TRAINING ON DIP: (1.45P.M-3.15P.M)

     Dr. D. Balasubramaniam, Associate Professor, GKMCET, covered topics

• Processing of brain image.
• High pass filter for edge detection.
• Low pass filter for image smoothening.
• Flow diagram of processing images and exhibiting different resolution of images.
• Pre-processing image after gray images and several processing techniques to obtain skeleton image, edge detection each part is taken and connected and viewed based on computer vision.
• Processing images FFT, Histogram graph.

DAY TWO -

Second day workshop was carried over by our guest speaker, Dr. U. Sabura Banu, Professor, B.S. Abdur Rahman Crescent Institute of Science and Technology, Chennai COVERING TOPICS

• Machine Learning is the process processing of data, while, deep learning is the process of processing with image to identify or classify a object data analytics modeling.
• From datas features are extracted and 80 percent is kept for training and creating model. 20 percent of sample is used for testing same model for accuracy.
• Clustering classification, Regression, Neural network app, curve filtering applications.
• Data are loaded using CSV file comma separated values machine learning are of 2 types.
• Unsupervised in which output is predicted example clustering.
• In supervised learning output is given regression example SVM.
• Linear and logistic regression polynomial regression.
• Identification flow type example in which sepal length, petal length width features are chosen which helps in better classification then training is done.
• Training model is done with 90 percent with accuracy in 30.38sec.
• Datasets are obtained from UCI repository and Kaggle website.
• MATLAB uses command Window
• Steps: Nprtool, load, Import data, hidden layers, train, Plot, seeing results, error histogram, confusion matrix overall accuracy.
• Gensim(net) for box (tool box) Simulink, sink display and run.

The afternoon session was covered with

MACHINE LEARNING FOR CLASSIFICATION NEURAL NETWORK   PATTERN RECOGNIZER   CLUSTRING SVM,NAÏVE BAYEE, CLASSIFICATION LEARNER APP, EEP LEARNING USING MATLAB PRETAINED ALEX NETWORK:   Latee the concluding session had certifcate distribution for the participats. the Particpants also gave their feed back on this workshop. The workshop concluded with vote of thanks to on and all and to the IEEE Control System Society - Madras Section.

---

---

---

---

DATE	DAY	TIMINGS	TOPICS	SPEAKER
26.09.2019	THRUSDAY	9.00A.M-10.00A.M	REGISTRATION
		10.00A.M-1.00P.M	MEDICAL IMAGE PROCESSING WITH PIXELS	Dr. B. Hema Kumar Assistant Professor, Pondicherry Engineering College
		1.00P.M-1.45P.M	LUNCH
		1.45P.M-3.15P.M	HANDS ON TRAINING ON DIP	Dr. D. Balasubramaniam Associate Professor, GKMCET
27.09.2019	FRIDAY	9.00A.M-10.30A.M	IMAGE PROCESSING USING MATLAB	Dr. U. Sabura Banu Professor, B.S. Abhur Rahman Crescent Institute of Science and Technology
		10.30A.M-10.45A.M	TEA BREAK
		10.45A.M-12.15P.M	STATISTICAL FEATURE EXTRACTION GEOMETRIC FEATURE EXTRACTION TEXTURE FEATURE EXTRACTION	Dr. U. Sabura Banu Professor, B.S. Abhur Rahman Crescent Institute of Science and Technology
		12.15P.M-1.00P.M	LUNCH
		1.00P.M-2.30P.M	MACHINE LEARNING FOR CLASSIFICATION NEURAL NETWORK     PATTERN RECOGNIZER     CLUSTRING SVM, NAÏVE BAYEE, CLASSIFICATION LEARNER APP, DEEP LEARNING USING MATLAB PRETAINED ALEX NETWORK	Dr. U. Sabura Banu Professor, B.S. Abhur Rahman Crescent Institute of Science and Technology
		2.30P.M-3.15P.M	CERTIFICATE DISTRIBUTION

---