Using Flipped Class-room to extend exercise time in Digital Image Processing

Author: dyrmann

Created: 2020-02-16 10:58pm

Edited: 2020-06-05 10:09am

Description:

The first requirement to be able to release time for supporting in-class exercises is to reduce the in-class lecture time. This means that large parts of the theory from the in-class lecture must be moved from the lesson (in class) to the preparation (out-of-class). The course is thereby partly transformed into a flipped classroom (FC) course. As a replacement for the traditional lecture, I will make videos that combine screencasts and pencasts, to replace the main parts of current lectures. Screencasts can be used for slides and graphic presentations, which today are presented using 'PowerPoint' slides. Screencasts can also be used for presenting code and examples. Fortunately, transforming these parts of the lecture to videos can easily be done, as this course is an algorithmic and programming course where the content is mostly digital. Pencasts can replace large parts of the blackboard-presentations, which consists primarily of presenting algorithms. The videos should be accompanied by quizzes that give the students feedback on their understanding of today's lesson, as well as provide inputs to the lecturer in the in-class teaching, thereby transforming it to just-in-time-teaching (JiTT).

When most of the theory is presented during the preparation time, it is necessary to allow the students to ask questions and get support in the preparation time. For this, a discussion board/chat-room is created, which can be used both before and after the lesson, and which the students are encouraged to actively contribute to.

With the freed-up time in-class, it is possible to focus the lecture on brief practical demonstrations, perspectives and use cases. This gives the students more time to work on the exercises, and the opportunity to get support while working on them.

After the lesson (out-of-class), the same discussion board can be used to ask questions related to the exercises, and other students, as well as the lecturer, can help to answer those questions.

Resources		Tasks		Supports
Before class
Textbook, screencasts and pencasts of the theory	→	Understanding theory and algorithms of today's lecture ↓
Quiz on LMS (eg. BlackBoard)	→	Students answer questions about today's topic ↓	←	Support from lecturer and peers on Discord or Riot
In class
Blackboard, slides, and answers from quizzes from the 'before class' activities		The lecturer presents today's topic with focus on practical use of the algorithms. ↓	←	Support from lecturer
Textbooks, demos, own notes		Practical exercise ↓	←	Support from lecturer
After class
Textbooks, demos, own notes		Practical exercise	←	Support from lecturer and peers on Discord or Riot

Additional information

Background:
“Digital Image Processing” is a course at the 6th semester at the bachelor level. It introduces algorithms and programming techniques to automatically detect objects in images.
Few topics in the course span two lessons, but most topics fill only one lesson.
All lessons are currently structured as follows:
1. Preparation (reading) (out of class)
2. Lecture (in class)
3. Exercises (out of class)
The reading (out-of-class) prepares the students for the lecture.
In the lecture, the theory from today's lesson is covered and concretised, after which the students themselves work actively with the topic in exercises.
In class, there is time to get started working on the exercises, but not completed during the four-lesson block devoted to the course each week. The students are therefore expected to work independently with the exercises after the lesson.
In the lesson, the students can ask the teacher about the theory for today's topic and exercises, but there is limited opportunity for getting help and feedback on the exercises before the next lesson, where the exercises are expected to have been completed already.
This learning design aims to increase the possibility of assistance and feedback on exercises during the lessons.