Course title
P0600300,Z3600200
Pattern Recognition

 takahashi masanobu Click to show questionnaire result at 2017
Course description
Pattern information is the information spreading spatially and/or temporally such as information related to five senses.
There is a great need for automated recognition of pattern information to help and/or substitute humans.
In this course, a series of processing is lectured for recognizing handwriting numerical characters. Exercises are also carried out to realize the processing by C-programing and evaluate the recognition rate.
Purpose of class
To understand a series of processing for recognizing handwriting numeral characters.
To acquire fundamental and practical knowledge through programing and experiments.
Goals and objectives
  1. To understand and memorize technical terms concerning pattern recognition.
  2. To understand and memorize a series of processing concerning character recognition.
  3. To correctly understand the meanings and features of the methods via experiments.
Language
Japanese
Class schedule

 Class schedule HW assignments (Including preparation and review of the class.) Amount of Time Required
1. Pre-processing 1: Segmentation Reference book 9 100minutes
Reviewal 90minutes
2. Pre-processing 1: Exercise on segmentation Report writing 190minutes
3. Pre-processing 2: Normalization (position and size) Reference book 8 70minutes
Reviewal 90minutes
4. Pre-processing 2: Exercise on normalization (position and size) Report writing 220minutes
5. Recognition 1: Pattern matching Reference book 11-1 70minutes
Reviewal 90minutes
6. Recognition 1: Exercise on pattern matching Report writing 220minutes
7. Feature extraction 1: Scaling down and smoothing Reference book 5 70minutes
Reviewal 90minutes
8. Feature extraction 2: Dimensional compression using KL transformation Reference book a-2-5 70minutes
Reviewal 90minutes
9. Feature extraction 2: Exercise on dimensional compression using KL transformation Report writing 220minutes
10. Recognition 2: Statistical recognition Reference book 12-1 70minutes
Reviewal 90minutes
11. Recognition 2: Exercise on statistical recognition Report writing 220minutes
12. Recognition 3: Neural networks Reference book 12-3-2 70minutes
Reviewal 90minutes
13. Recognition 3: Exercise on neural networks Report writing 220minutes
14. Deep learning and examination Reference 12-3-2[3] 20minutes
Reviewal of all contents 200minutes
Total. - - 2660minutes
Relationship between 'Goals and Objectives' and 'Course Outcomes'

 Report Examination Q&A Total.
1. 10% 16% 1% 27%
2. 10% 16% 1% 27%
3. 30% 16% 0% 46%
Total. 50% 48% 2% -
Evaluation method and criteria
Reports (6 times), Examination and Q&A are used for evaluation.
Textbooks and reference materials
Handouts are used. Reference book is "Digital image processing" by CG-ARTS.
Prerequisites
It is desirable to have studied digital image processing.
Office hours and How to contact professors for questions
  • Basically Friday afternoon.
Relation to the environment
Non-environment-related course
Regionally-oriented
Non-regionally-oriented course
Development of social and professional independence
  • Course that cultivates a basic problem-solving skills
  • Course that cultivates an ability for utilizing knowledge
  • Course that cultivates a basic self-management skills
Active-learning course
About half of the classes are interactive
Last modified : Wed Oct 17 06:57:26 JST 2018