No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Machine Vision Algorithm Training Course Construction with PBL
Abstract
Interdisciplinary integration of theory and practice is imperative as a course requirement in emerging engineering education, and in the public elective course "Machine Vision Algorithm Training". Considering the entire teaching process, including pre-training, in-training, and post-training, this paper discusses the course construction and content in detail in terms of project-based learning (PBL). The PBL teaching approach and evaluation methods are described in detail through a comprehensive face recognition training case based on a convolutional neural network (CNN) and Raspberry Pi. Through project design training from shallower to deeper, interdisciplinary integration of theory and practice is cultivated, stimulating interest in course study. The results demonstrate that PBL teaching improves the engineering application and innovative abilities of students.
... In some instances, the literature can be categorized into both categories. The common theme for literature that implements machine vision elements within the project-based learning environment is that students are engaged due to the perceived utility of machine vision [2] and the challenge of applying and mastering the technique [7]. ...
Machine vision is even more critical for sorting workpieces in quality control of automated production systems for Industry 4.0. Use the image processing system principle to inspect the workpiece with a high-speed vision sensor camera. It results in high accuracy of sorting performance and can replace labor in automated production processes. The learning process of vision sensors used for inspecting workpieces with industrial cameras for mechatronics and robotics engineering students is concerned. Most barriers need more teaching materials due to the relatively high cost of the training kit. That is causing students' lack of knowledge and practical skills during the professional training experience in the workplace. This study proposed designing a sorting system using a machine vision (SSMV) training kit for mechatronics and robotics engineering students. It can be a learning tool to promote the students' learning in related fields, automated manufacturing systems, and engineering education.
This paper introduces project-based learning (PBL) features for developing technological, curricular, and pedagogical supports to engage students in computational thinking (CT) through modeling. CT is recognized as the collection of approaches that involve people in computational problem solving. CT supports students in deconstructing and reformulating a phenomenon such that it can be resolved using an information-processing agent (human or machine) to reach a scientifically appropriate explanation of a phenomenon. PBL allows students to learn by doing, to apply ideas, figure out how phenomena occur and solve challenging, compelling and complex problems. In doing so, students take part in authentic science practices similar to those of professionals in science or engineering, such as computational thinking. This paper includes 1) CT and its associated aspects, 2) The foundation of PBL, 3) PBL design features to support CT through modeling, and 4) a curriculum example and associated student models to illustrate how particular design features can be used for developing high school physical science materials, such as an evaporative cooling unit to promote the teaching and learning of CT.
In this paper, a novel lightweight incremental class learning algorithm for live image recognition is presented. It features a dual memory architecture and is capable of learning formerly unknown classes as well as conducting its learning across multiple instances at multiple locations without storing any images. In addition to tests on the ImageNet dataset, a prototype based upon a Raspberry Pi and a webcam is used for further evaluation: The proposed algorithm successfully allows for the performant execution of image classification tasks while learning new classes at several sites simultaneously, thereby enabling its application to various industry use cases, e.g. predictive maintenance or self-optimization.
Engineering Exploration is a first-year engineering course designed with enduring outcomes of enhancing problem-solving skills, multidisciplinary engineering skills and promoting team work. In order to address all the enduring outcomes of this course, traditional pedagogical technique was found to be challenging which motivated the authors to use Project Based Learning (PBL) approach. This paper evaluates the elements of a first-year course titled engineering exploration against gold standard PBL framework (defined by Buck institute for Education) there by identifies and proposes the essentials to be considered while designing a first year PBL course.
italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Contribution:
This article shares the learning ecosystem of a project-based embedded systems course, identifying course elements that support self-directed learning and how assignments guide students toward becoming adaptive experts.
Background:
The technology advances while the fundamentals of electrical engineering remain static. Educators can increasingly prepare students to identify what they need to know to solve problems and avail themselves of resources to learn. This article seeks to further understand ways that a project-based learning approach in an undergraduate embedded systems course can facilitate students’ self-directed learning.
Research Question:
In what ways can a project-based learning approach in an undergraduate embedded systems course facilitates the self-directed learning amongst students?
Methodology:
This article, conducted in the context of an existing embedded systems design (ESD) course, relied on interviews of students, teaching assistants, and faculty along with document analysis and a mixed inductive–deductive thematic analysis.
Findings:
A learning ecology of the course is presented. This includes descriptions of space and facilities that influence student motivation, means by which the pedagogical intent of the instructor impacts the student experience, how the course builds on project-based learning knowledge, how the content is distributed using knowledge sharing, how Making supported the ecosystem, how students and instructor occupy similar roles, how the curricular design process was conducted, and how the open ecology promotes student self-direction.
Distance education has been widely used in universities due to the outbreak of Covid-19. SPOCs (Small Private Online Courses) have high enrollment and completion rates in higher education due to their ability to provide students with personalized deep learning. This study aims to investigate the learning effect of SPOCs and further explores the reasons that affect its effectiveness. The study found that: a) there is a certain difference in the learning effect between SPOCs and face-to-face courses; b) the learning effect in SPOCs can be effectively improved by enhancing the interest in learning, and c) the teacher’s internal and external assistance can significantly improve the learning effect. The study suggested that the “Hard-Pure” SPOCs needs more detailed course design, enhancing students’ motivations and learning interests in SPOC courses, and strengthening teachers’ assistance to students in and out of class.
The education systems around the world have had to adapt quickly to find ways to offer university programs remotely in the face of the Covid 19 pandemic. Without the proper preparation and sufficient knowledge, instructors learned to teach online during the Winter 2020 semester to allow students to complete their courses when campuses suddenly closed as a safety measure. Normally many weeks of preparation would have been necessary to redesign a course to be efficiently offered online. This paper studies some aspects of the transition from Face to Face teaching in the Fall 2019 semester to Face to face / Online teaching in the Winter of 2020 (beginning of the pandemic) to completely Online teaching in the Fall 2020. This transition is examined in a Managerial Analytics course offered in the first semester of an MBA program at a Canadian University. A survey administered at the end of each semester reveals different levels of students’ anxiety, modification in the communication tools utilized, changes in intensity of weekly study hours and expected recollection of the material learned in the course, a year after completion. Additional variations are also observed by gender.
Upon graduation, a computer science student should have a good understanding of the current technology and have the soft skills necessary to secure a position in industry. Considering that typical computers and even the common smartphone are multicore, students should be skilled in parallel programming. Integrating parallel programming and soft skills within courses can help educate students on these essential skills. Our goal is to explore the effectiveness of using Project Based Learning (PBL) to teach these skills when classes are at content capacity. We divide 247 students into 51 diverse groups and assigned five projects, each of two-week duration. We use pre- and post-surveys to measure growth and found that incorporating PBL has a significant effect on the students' parallel programming and soft skills. We show that through teamwork, students collaboratively learn and apply fundamental parallel programming and soft skills without direct guidance, thus demonstrating the effectiveness of PBL. The implementation was conducted in a course that does not traditionally teach parallel programming concepts, but with the use of a PBL approach, students were able to acquire this new knowledge.
Implementing artificial intelligence (AI) in the Internet of Things (IoT) involves a move from the cloud to the heterogeneous and low-power edge, following an urgent demand for deploying complex training tasks in a distributed and reliable manner. This work proposes a self-aware distributed deep learning (DDL) framework for IoT applications, which is applicable to heterogeneous edge devices aiming to improve adaptivity and amortize the training cost. The self-aware design including the dynamic self-organizing approach and the self-healing method enhances the system reliability and resilience. Three typical edge devices are adopted with cross-platform Docker deployment: Personal Computers (PC) for general computing devices, Raspberry Pi 4Bs (Rpi) for resource-constrained edge devices, and Jetson Nanos (Jts) for AI-enabled edge devices. Benchmarked with ResNet-32 on CIFAR-10, the training efficiency of tested distributed clusters is increased by 8.44× compared to the standalone Rpi. The cluster with 11 heterogeneous edge devices achieves a training efficiency of 200.4 images/sec and an accuracy of 92.45%. Results prove that the self-organizing approach functions well with dynamic changes like devices being removed or added. The self-healing method is evaluated with various stabilities, cluster scales, and breakdown cases, testifying that the reliability can be largely enhanced for extensively distributed deployments. The proposed DDL framework shows excellent performance for training implementation with heterogeneous edge devices in IoT applications with high-degree scalability and reliability.
One of the fast-growing areas of deep learning using artificial intelligence is computer vision, becoming increasingly popular. It is a growing field of research that seeks to create techniques to help computers “see” and recognize the information of digital images for instance videos and photographs. Object detection is a computer vision method that enables us to recognize objects in an image or video and locate them. This article describes an efficient shape-based object identification method and its displacement in real-time using OpenCV library of programming roles mostly targeted at computer vision and Raspberry Pi with camera module.
Internet of Things (IoT) has become a very popular area of research due to its large-scale implementation and challenges. However, security is the key concern while witnessing the rapid growth in its size and applications. It is a tedious task to individually put security mechanisms in each IoT device and update it as per newer threats. Moreover, machine learning models can best utilize the colossal amount of data generated by IoT devices. Therefore, many Deep Learning (DL) based mechanisms have been proposed to detect attacks in IoT. However, the existing security mechanisms addressed limited attacks, and they used limited and outdated datasets for evaluations. This paper presents a novel security framework and an attack detection mechanism using a Deep Learning model to fill in the gap, which will efficiently detect malicious devices. The proposed mechanism uses a Convolution Neural Network (CNN) to extract the accurate feature representation of data and further classifies those by Long Short-Term Memory (LSTM) Model. The dataset used in the experimental evaluation is from twenty Raspberry Pi infected IoT devices. The accuracy of the empirical study for attack detection is 96 percent. In addition, it is observed that the proposed model outperformed various recently proposed DL-based attack detection mechanisms.
The paper discusses the effective organization of students’ project-based learning activities in an engineering university. The educational process modernization involves the introduction of new teaching approaches related to the development of additional competencies of students which allows them to implement complex projects, consequently increasing the competitiveness of university graduates in Russia. Numerous studies on project-based learning show it as the most suitable means of achieving effective competence-based education. The research proposes that the dissemination students’ project-based learning activities in an engineering university will be effective when the following pedagogical conditions are implemented: targeted development of students’ project-based learning activities; using student-student interaction in the students’ project-based learning activities; the readiness of the university faculty to organize and implement students’ project-based learning activities. Criteria for students’ project-based learning competencies are basic and in-depth knowledge in engineering field; professional competencies and personal qualities, interpersonal skills, ability to initiate, develop and implement projects. The paper is based on some successful examples of effective projects done by engineering students.
Deep learning has made essential contributions to classification and detection tasks applied to precision agriculture; however, it is vitally important to move towards an adoption of these techniques and algorithms through low-cost and low-consumption devices for daily use in crop fields. In this paper, we present the training and evaluation of four recent Convolutional Neural Networks models for the classification of diseases in tomato leaves. A subset of the Plantvillage dataset consisting of 18,160 RGB images has been divided into ten classes for transfer learning. The selected models have depthwise separable convolution architecture for application in low-power devices. Evaluation and analysis quantitatively and qualitatively is performed via quality metrics and saliency maps. Finally, an implementation on the Raspberry Pi 4 microcomputer with a graphical user interface is developed.
