TGS Salt Identification Challenge

Slides:



Advertisements
Similar presentations
Patient information extraction in digitized X-ray imagery Hsien-Huang P. Wu Department of Electrical Engineering, National Yunlin University of Science.
Advertisements

Final Project: Project 9 Part 1: Neural Networks Part 2: Overview of Classifiers Aparna S. Varde April 28, 2005 CS539: Machine Learning Course Instructor:
Face Processing System Presented by: Harvest Jang Group meeting Fall 2002.
An Example of Course Project Face Identification.
Mobile Image Processing
Deep Convolutional Nets
Neural networks (2) Reminder Avoiding overfitting Deep neural network Brief summary of supervised learning methods.
Chapter 11 – Neural Nets © Galit Shmueli and Peter Bruce 2010 Data Mining for Business Intelligence Shmueli, Patel & Bruce.
Joshua Myrans, Prof. Zoran Kapelan & Prof. Richard Everson Automating Detection of faults in small wastewater pipes, using CCTV footage Challenge Wastewater.
Anomaly Detection in Data Science
Combining Models Foundations of Algorithms and Machine Learning (CS60020), IIT KGP, 2017: Indrajit Bhattacharya.
Summary of “Efficient Deep Learning for Stereo Matching”
Deep Learning Amin Sobhani.
Neural Networks for Quantum Simulation
Computer Science and Engineering, Seoul National University
Convolutional Neural Fabrics by Shreyas Saxena, Jakob Verbeek
The Problem: Classification
Many slides and slide ideas thanks to Marc'Aurelio Ranzato and Michael Nielson.
Applications of Deep Learning and how to get started with implementation of deep learning Presentation By : Manaswi Advisor : Dr.Chinmay.
Depth estimation and Plane detection
Machine Learning I & II.
Deep Learning Libraries
Object Detection with Bootstrapping Carlos Rubiano Mentor: Oliver Nina
Ajita Rattani and Reza Derakhshani,
Inception and Residual Architecture in Deep Convolutional Networks
CAMELYON16 Challenge Matt Berseth, NLP Logix Jacksonville FL
CS 188: Artificial Intelligence
CS 698 | Current Topics in Data Science
CS 698 | Current Topics in Data Science
R-CNN region By Ilia Iofedov 11/11/2018 BGU, DNN course 2016.
Project Implementation for ITCS4122
Enhanced-alignment Measure for Binary Foreground Map Evaluation
ECE 599/692 – Deep Learning Lecture 6 – CNN: The Variants
Deep Learning Convoluted Neural Networks Part 2 11/13/
Master’s Thesis defense Ming Du Advisor: Dr. Yi Shang
Fully Convolutional Networks for Semantic Segmentation
CSC 578 Neural Networks and Deep Learning
Machine Learning Platform Life-Cycle Management
Attention-based Caption Description Mun Jonghwan.
Image Classification.
Jason furmanek Kris murphy IBM
Technical Advisory Committee
cs540 - Fall 2016 (Shavlik©), Lecture 20, Week 11
ECE 599/692 – Deep Learning Lecture 4 – CNN: Practical Issues
Smart Robots, Drones, IoT
Technical Advisory Committee
CSC 578 Neural Networks and Deep Learning
A Proposal Defense On Deep Residual Network For Face Recognition Presented By SAGAR MISHRA MECE
Neural Networks Geoff Hulten.
YOLO-LITE: A Real-Time Object Detection Web Implementation
Neural Speech Synthesis with Transformer Network
Vinit Shah, Joseph Picone and Iyad Obeid
Tuning CNN: Tips & Tricks
Object Tracking: Comparison of
Semantic Similarity Detection
Coding neural networks: A gentle Introduction to keras
Deep Learning for Plant Stress Phenotyping: Trends and Future Perspectives  Asheesh Kumar Singh, Baskar Ganapathysubramanian, Soumik Sarkar, Arti Singh 
Keras.
Mean.
Gene Structure Prediction Using Neural Networks and Hidden Markov Models June 18, 권동섭 신수용 조동연.
Deep Learning Libraries
CS295: Modern Systems: Application Case Study Neural Network Accelerator Sang-Woo Jun Spring 2019 Many slides adapted from Hyoukjun Kwon‘s Gatech “Designing.
VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION
Rohan Yadav and Charles Yuan (rohany) (chenhuiy)
Debasis Bhattacharya, JD, DBA University of Hawaii Maui College
Multi-UAV to UAV Tracking
CSC 578 Neural Networks and Deep Learning
LHC beam mode classification
Introduction Face detection and alignment are essential to many applications such as face recognition, facial expression recognition, age identification,
on Road Signs & Face Detection
Presentation transcript:

TGS Salt Identification Challenge —— Zheming Wang Course Work of CS539 in Fall 2018

Introduction Salt deposits below the surface is a big obstacle for oil and gas drillers. To detect them, expert human interpretation is still a must nowadays. The goal of this project is to tell the salt region with given seismic image as follow:

Data and method The data comes from the TGS challenge on Kaggle. The picture size is 101*101, while the train and test sets have sizes of 4,000 and 18,000. The tool kit I use is Keras with TensorFlow as backend in python. The code I used mainly referred to some kernels on Kaggle. I keep the metric and data transforming part unchanged, revised the model structure, and wrote some code to make the training resumable and automatic on Euler.

Basic Summary It is a typical image segmentation problem and there is a specific metric called Intersection over Union (IoU). In this project, I start from a baseline U-net model with IoU of 0.675(0.700) , ranking at 81%. After trying different methods and do lots of fine tuning, I get a 2-stage Res-Unet model with data augmentation, cross validation and voting, which obtains IoU of 0.839(0.858) , ranking at 10%.

Model Structure: U-net (baseline)

Model Structure: Runet-revised (No resize)

Result 1: model choosing description test score in leaderboard private public position U-net baseline 0.700 0.675 80.7% U-net - improved use greyscale 0.753 0.722 72.3% Res-U-net (Runet) add ResNet; add threshold 0.818 0.797 47.9% Runet-depth add depth as a channel 0.809 0.782 55.5% Runet-csum add cum-sum as a channel 0.813 0.793 50.6% Runet-revised keep input size 101*101 unchanged 0.829 0.800 46.3% General setting Batch size: 32 Epochs: 200 Early Stop : 10 Lr schedule : 1e-2/ 8/ 0.5/ 1e-5 Monitor : val_iou * May differ a little per model.

test score in leaderboard Result 2: fine-tuning Model description test score in leaderboard private public position Runet-lovasz 50 + 100 (baseline) 0.8323 0.8005 45.8% Runet-lovasz - tuned 100 + 200 0.8457 0.8171 32.7% cross validation 3 500 + 300 best-cv 0.8423 0.8180 31.9% combination - voting 0.8505 0.8264 21.1% cross validation 5-1 Augmentation; 300 + 300 best-cv 0.8456 0.8236 26.6% cross valiadation 5-2 Augmentation; 150 + 200 best-cv 0.8539 0.8334 14.0% 0.8581 0.8352 12.0% combination - best3 best 3 above 0.8582 0.8393 10.2%

Lessons Learned in Project 1. Keras has some problems when resume a saving model. After reading source code, I wrote a patch for it. https://github.com/wzmJimmy/MyKerasPatch 2. Parallelism did not accelerate single model training. I found this fact when trying Keras mult-gpu class. Then I changed to train multiple cross-validation like models at the same time. 3. In Data Science, the model choosing is a core problem, while the data preprocessing and hyper-parameter tuning can also make a great difference but need lots of experience and trails.

test score in leaderboard Further improvement Other techniques: Transfer learning: using other network as an encoder. Optimizer choosing: Adam, momentum, normal SGD. Feature engineering: tell zero-or-some salt first. …… Model description test score in leaderboard private public position model with X-ception as encoder, different optimizer, tricks like pseudo-labeling and Stochastic weight averaging. 0.8771 0.8547 4.2%

Thank you for listening.