scholarly journals Research of Metric Learning-Based Method for Person Re-Identification by Intelligent Computer Vision Technology

2021 ◽  
Vol 2083 (4) ◽  
pp. 042013
Author(s):  
Qiyu Rao
Keyword(s):  
Binary Classification ◽  
Metric Learning ◽  
Training Data ◽  
Identification Accuracy ◽  
Learning Method ◽  
Learning Methods ◽  
The Difference ◽  
Multi Class Classification ◽  
Fitting Problem ◽  
Metric Model

Abstract Person re-identification technology aims to establish an efficient metric model for similarity distance measurement of pedestrian images. Candidate images captured by different camera views are ranked according to their similarities to the target individual. However, the metric learning-based method, which is commonly used in similarity measurement, often failed in person re-identification tasks due to the drastic variations in appearance. The main reason for its low identification accuracy is that the metric learning method is over-fitting to the training data. Several types of metric learning methods which differ from each other by the distribution of sample pairs were summarized in this article for analysing and easing the metric learning methods’ over-fitting problem. Three different metric learning methods were tested on the VIPeR dataset. The distributions of the distance of the positive/negative training/test pairs are displayed to demonstrate the over-fitting problem. Then, a new metric model was proposed by combining the thoughts of binary classification and multi-class classification. Related verification experiments were conducted on VIPeR dataset. Besides, the semi-supervised metric learning approach was introduced to alleviate the over-fitting problem. The experimental results reflect gap between training pairs and test pairs in the metric subspace. Therefore, reducing the difference between training data and test data is a promising way to improve the identification accuracy of metric learning method.

scholarly journals Pengaruh problem-based learning terhadap hasil belajar ditinjau dari motivasi belajar PLC di SMK

2013 ◽  
Vol 3 (2) ◽  
Author(s):  
Bekti Wulandari ◽  
Herman Dwi Surjono
Keyword(s):  
Learning Outcomes ◽  
Problem Based Learning ◽  
Learning Outcome ◽  
Motivation To Learn ◽  
Learning Method ◽  
Learning Methods ◽  
Plc Control ◽  
The Subject ◽  
The Difference ◽  
Interactional Effect

Penelitian ini bertujuan untuk mengetahui: (1) perbedaan hasil belajar siswa pada mata pelajaran pemrograman sistem kendali PLC antara siswa yang diajar dengan metode PBL dengan siswa yang diajar dengan metode demonstrasi, (2) pengaruh interaksi antara metode PBL dan metode demonstrasi dengan motivasi belajar terhadap hasil belajar siswa, (3) perbedaan hasil belajar siswa antara siswa yang diajar dengan metode PBL dengan yang diajar dengan metode demonstrasi ditinjau dari motivasi belajar. Penelitian ini merupakan penelitian eksperimen semu dengan desain faktorial yang dilakukan dengan memberikan perlakuan metode pembelajaran. Analisis data dalam penelitian ini menggunakan uji-t dan ANAVA dengan program SPSS 16. Hasil penelitian menunjukkan bahwa: (1) terdapat perbedaan hasil belajar antara siswa yang diajar dengan metode PBL dengan yang diajar dengan metode demonstrasi, (2) tidak terdapat pengaruh interaksi antara metode PBL dan demonstrasi dengan motivasi belajar terhadap hasil belajar, (3) terdapat perbedaan hasil belajar antara siswa yang diajar dengan metode PBL dengan yang diajar dengan metode demonstrasi ditinjau dari motivasi tinggi dan rendah. THE EFFECT OF PROBLEM-BASED LEARNING ON THE LEARNING OUTCOMES SEEN FROM MOTIVATION ON THE SUBJECT MATTER OF PLC IN SMKAbstractThis study aimed to determine: (1) the difference in learning outcomes in the subject of PLC control system programming between the students taught using the PBL method and those taught using the demonstration method, (2) the effect of the interaction among the PBL method and demonstration learning method with motivation on the learning outcomes, (3) difference in students’ learning outcomes between the students taught using the PBL method and those taught using the demonstration learning method in terms of motivation to learn. This study was a quasiexperimental study with a factorial design done by giving treatment in learning methods. The techniques of data analysis were T test and analysis of variance (ANOVA) with SPSS 16. The results show that: (1) there is a difference in learning outcomes between the students taught using the PBL method compared to those taught using the demonstration learning method, (2) there is no interactional effect between PBL and demonstration learning methods with motivation on the students’ learning outcome, (3) there is a difference in the learning outcome between students taught using the PBL method and those taught using the demonstration learning method in terms of students with high and low motivation.

scholarly journals Reconstruction of Simulated Magnetic Resonance Fingerprinting Using Accelerated Distance Metric Learning

2020 ◽  
Author(s):  
Elmira Yazdan ◽  
Sajjad Aghabozorgi Sahaf ◽  
Hamidreza Saligheh Rad
Keyword(s):  
Magnetic Resonance ◽  
Metric Learning ◽  
Tissue Response ◽  
Training Data ◽  
Distance Metric Learning ◽  
Learning Method ◽  
Distance Metric ◽  
Matching Algorithm ◽  
Correlation Algorithm ◽  
Space Data

Purpose: Magnetic Resonance Fingerprinting (MRF) is a novel framework that uses a random acquisition to acquire a unique tissue response, or fingerprint. Through a pattern-matching algorithm, every voxel-vise fingerprint is matched with a pre-calculated dictionary of simulated fingerprints to obtain MR parameters of interest. Currently, a correlation algorithm performs the MRF matching, which is time-consuming. Moreover, MRF suffers from highly undersampled k-space data, thereby reconstructed images have aliasing artifact, propagated to the estimated quantitative maps. We propose using a distance metric learning method as a matching algorithm and a Singular Value Decomposition (SVD) to compress the dictionary, intending to promote the accuracy of MRF and expedite the matching process. Materials and Methods: In this investigation, a distance metric learning method, called the Relevant Component Analysis (RCA) was used to match the fingerprints from the undersampled data with a compressed dictionary to create quantitative maps accurately and rapidly. An Inversion Recovery Fast Imaging with Steady-State (IR-FISP) MRF sequence was simulated based on an Extended Phase Graph (EPG) on a digital brain phantom. The performance of our work was compared with the original MRF paper. Results: Effectiveness of our method was evaluated with statistical analysis. Compared with the correlation algorithm and full-sized dictionary, this method acquires tissue parameter maps with more accuracy and better computational speed. Conclusion: Our numerical results show that learning a distance metric of the undersampled training data accompanied by a compressed dictionary improves the accuracy of the MRF matching and overcomes the computation complexity.

scholarly journals Pembelajaran Metode Drill Untuk Meningkatkan Hasil Belajar Passing Bawah Pada Permainan Bolavoli Siswa SMPN 2 Mare Kabupaten Bone

2020 ◽  
Vol 5 (1) ◽  
pp. 1
Author(s):  
Andi Amry Yahya ◽  
Sufitriono Sufitriono
Keyword(s):  
Learning Outcomes ◽  
Junior High School ◽  
Random Sampling ◽  
Good Effect ◽  
Learning Method ◽  
Average Increase ◽  
Learning Methods ◽  
Significance Level ◽  
Average Value ◽  
The Difference

 Drill Learning Method To Improve Lower Passing Learning Outcomes in Volleyball at SMPN 2 Mare BoneThis study aims to determine the effect of the Drill learning method and the conventional learning method on the learning outcomes of passing under the volleyball game of students of SMPN 2 Mare Bone District. This research method is a type of experimental research chosen by random sampling and then given a pretest and posttest with a population of all students of SMPN 2 Mare Bone District. Next class VIIIA and class VIIIB were chosen to be the sample of this study, each numbering 30 students and then divided into two groups using ordinal paired maching. data analysis techniques used t test with a significance level of 0.05.The results of this study show there is a significant effect of conventional learning methods with an average increase of 7.93 with a significant level of 0.05. There is a significant effect of the Drill learning method with an average increase of 9.57 with a significant level of 0.05. There is a difference of influence between Conventional learning methods and Drill learning methods by showing differences or the difference in the average value of 1.633. thus showing that the Drill learning method has a very good effect compared to the conventional learning method.Keywords: Drill method, student learning outcomes, under pass, volleyball, junior high school Penelitian ini bertujuan untuk mengetahui pengaruh antara metode pembelajaran Drill dan metode pembelajaran konvensional terhadap hasil belajar passing bawah permainan bola voli siswa SMPN 2 Mare Kabupaten Bone. Metode penelitian ini merupakan jenis penelitian eksperimen yang dipilih secara random sampling kemudian diberi pretest dan posttest dengan populasi semua siswa SMPN 2 Mare Kabupaten Bone. Selanjutnya dipilih kelas VIIIA dan kelas VIIIB untuk menjadi sampel penelitian ini yang jumlahnya masing – masing 30 orang siswa kemudian dibagi menjadi dua kelompok dengan menggunakan maching ordinal paired. teknik analisis data yang digunakan uji t dengan taraf signifikan 0.05. Hasil penelitian ini menujukkan ada pengaruh yang signifikan metode pembelajaran konvensional dengan  peningkatan rata rata 7,93 dengan taraf signifikan 0,05. Ada pengaruh yang signifikan metode pembelajaran Drill dengan peningkatan rata rata 9.57 dengan taraf signifikan 0,05. Ada perbedaan pengaruh antara metode pembelajaran Konvensional den metode pembelajaran Drill dengan menunjukkan perbedaan atau selisih nilai rata rata 1,633. sehingga menunjukkan metode pembelajaran Drill memiliki pengaruh yang sangat baik dibanding metode pembelajaran Konvensional.Kata kunci: Metode Drill, hasil belajar siswa, pasing bawah, bola voli, SMP

scholarly journals An Unknown Radar Emitter Identification Method Based on Semi-Supervised and Transfer Learning

Algorithms ◽  
2019 ◽  
Vol 12 (12) ◽  
pp. 271 ◽  
Author(s):  
Yuntian Feng ◽  
Guoliang Wang ◽  
Zhipeng Liu ◽  
Runming Feng ◽  
Xiang Chen ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Training Data ◽  
Identification Accuracy ◽  
Support Vector ◽  
Learning Method ◽  
Target Domain ◽  
Identification Method ◽  
Svm Model ◽  
Testing Data ◽  
Task Simulation

Aiming at the current problem that it is difficult to deal with an unknown radar emitter in the radar emitter identification process, we propose an unknown radar emitter identification method based on semi-supervised and transfer learning. Firstly, we construct the support vector machine (SVM) model based on transfer learning, using the information of labeled samples in the source domain to train in the target domain, which can solve the problem that the training data and the testing data do not satisfy the same-distribution hypothesis. Then, we design a semi-supervised co-training algorithm using the information of unlabeled samples to enhance the training effect, which can solve the problem that insufficient labeled data results in inadequate training of the classifier. Finally, we combine the transfer learning method with the semi-supervised learning method for the unknown radar emitter identification task. Simulation experiments show that the proposed method can effectively identify an unknown radar emitter and still maintain high identification accuracy within a certain measurement error range.

Knowledge Management as an E-Learning Tool

2011 ◽  
pp. 376-384
Author(s):  
Javier Andrade ◽  
Juan Ares ◽  
Rafael García ◽  
Santiago Rodríguez ◽  
Maria Seoane ◽  
...  
Keyword(s):  
Communication Networks ◽  
Teaching Method ◽  
Educational Process ◽  
Learning Method ◽  
Learning Methods ◽  
Educational Methods ◽  
E Learning ◽  
The Difference ◽  
New Vision ◽  
Pupil Learning

The goal of educational methods is to allow the pupil the acquisition of knowledge. Even so, the way in which this aim is pursued originates four different currents of methods sorted by two criteria: (1) who leads the educational process and (2) requirement of pupil physical attendance. Regarding the former criterion, the process may be conducted either by the teacher—Teaching-Oriented Process—or by the pupil—Learning-Oriented Process. Obviously, both processes have the same aim: the interiorization and comprehension of knowledge by the pupil. But the difference between them is based on the distinctive procedure followed in each case to achieve the common goal. Regarding the second criterion, the methods may or may not require pupil attendance. Bearing in mind this classification, four different types of educational methods could be described: 1. Teaching Method: This includes the already known classic educational methods, the Conductivity Theory (Good & Brophy, 1990) being the foremost one. This method is characterized by the fact that the teacher has the heavier role during education—the transmission of knowledge. 2. E-Teaching Method: This second type comes from the expansion and popularity of communication networks, especially the Internet. This method brings the teacher to the physical location of the pupil; one of its most important representative elements is the videoconference. 3. Learning Method: This constitutes a new vision of the educational process, since the teacher acts as a guide and reinforcement for the pupil. The educational process has the heavier role in this method. In other words, the teacher creates a need for learning and afterwards provides the pupil with the necessary means in order to fill these created requests. Piaget Constructionist Theory is one of the most remarkable methods for this (Piaget, 1972, 1998). 4. E-Learning Method: This method is supported both by learning methods and by the expansion of communication networks in order to facilitate access to education with no physical or temporal dependence from pupil or teacher. As in learning methods, the pupil, not the teacher, is the one who sets the learning rhythm.

Knowledge Management as an E-Learning Tool

2006 ◽  
pp. 381-388
Author(s):  
Javier Andrade ◽  
Juan Ares ◽  
Rafael García ◽  
Santiago Rodríguez ◽  
María Seoane ◽  
...  
Keyword(s):  
Communication Networks ◽  
Teaching Method ◽  
Educational Process ◽  
Learning Method ◽  
Learning Methods ◽  
Educational Methods ◽  
E Learning ◽  
The Difference ◽  
New Vision ◽  
Pupil Learning

The goal of educational methods is to allow the pupil the acquisition of knowledge. Even so, the way in which this aim is pursued originates four different currents of methods sorted by two criteria: (1) who leads the educational process and (2) requirement of pupil physical attendance. Regarding the former criterion, the process may be conducted either by the teacher—Teaching-Oriented Process—or by the pupil—Learning-Oriented Process. Obviously, both processes have the same aim: the interiorization and comprehension of knowledge by the pupil. But the difference between them is based on the distinctive procedure followed in each case to achieve the common goal. Regarding the second criterion, the methods may or may not require pupil attendance. Bearing in mind this classification, four different types of educational methods could be described: 1. Teaching Method: This includes the already known classic educational methods, the Conductivity Theory (Good & Brophy, 1990) being the foremost one. This method is characterized by the fact that the teacher has the heavier role during education—the transmission of knowledge. 2. E-Teaching Method: This second type comes from the expansion and popularity of communication networks, especially the Internet. This method brings the teacher to the physical location of the pupil; one of its most important representative elements is the videoconference. 3. Learning Method: This constitutes a new vision of the educational process, since the teacher acts as a guide and reinforcement for the pupil. The educational process has the heavier role in this method. In other words, the teacher creates a need for learning and afterwards provides the pupil with the necessary means in order to fill these created requests. Piaget Constructionist Theory is one of the most remarkable methods for this (Piaget, 1972, 1998). 4. E-Learning Method: This method is supported both by learning methods and by the expansion of communication networks in order to facilitate access to education with no physical or temporal dependence from pupil or teacher. As in learning methods, the pupil, not the teacher, is the one who sets the learning rhythm.

scholarly journals A new approach to image classification based on a deep multiclass AdaBoosting ensemble

2020 ◽  
Vol 10 (5) ◽  
pp. 4872
Author(s):  
Hasan Asil ◽  
Jamshid Bagherzadeh
Keyword(s):  
Deep Learning ◽  
Error Rate ◽  
Hybrid Methods ◽  
Learning Method ◽  
Learning Methods ◽  
Data Set ◽  
Boosting Algorithm ◽  
Multi Class Classification ◽  
Classification Of Images

In recent years, deep learning methods have been developed in order to solve the problems. These methods were effective in solving complex problems. Convolution is one of the learning methods. This method is applied in classifying and processing of images as well. Hybrid methods are another multi-component machine learning method. These methods are categorized into independent and dependent types. Ada-Boosting algorithm is one of these methods. Today, the classification of images has many applications. So far, several algorithms have been presented for binary and multi-class classification. Most of the above-mentioned methods have a high dependence on the data. The present study intends to use a combination of deep learning methods and associated hybrid methods to classify the images. It is presumed that this method is able to reduce the error rate in images classification. The proposed algorithm consists of the Ada-Boosting hybrid method and bi-layer convolutional learning method. The proposed method was analyzed after it was implemented on a multi-class Mnist data set and displayed the result of the error rate reduction. The results of this study indicate that the error rate of the proposed method is less than Ada-Boosting and convolution methods. Also, the network has more stability compared to the other methods.

scholarly journals On the Learning Property of Logistic and Softmax Losses for Deep Neural Networks

2020 ◽  
Vol 34 (04) ◽  
pp. 4739-4746
Author(s):  
Xiangrui Li ◽  
Xin Li ◽  
Deng Pan ◽  
Dongxiao Zhu
Keyword(s):  
Neural Networks ◽  
Visual Recognition ◽  
Binary Classification ◽  
Training Data ◽  
Necessary Condition ◽  
Deep Convolutional Neural Networks ◽  
Target Class ◽  
Benchmark Datasets ◽  
Negative Class ◽  
Multi Class Classification

Deep convolutional neural networks (CNNs) trained with logistic and softmax losses have made significant advancement in visual recognition tasks in computer vision. When training data exhibit class imbalances, the class-wise reweighted version of logistic and softmax losses are often used to boost performance of the unweighted version. In this paper, motivated to explain the reweighting mechanism, we explicate the learning property of those two loss functions by analyzing the necessary condition (e.g., gradient equals to zero) after training CNNs to converge to a local minimum. The analysis immediately provides us explanations for understanding (1) quantitative effects of the class-wise reweighting mechanism: deterministic effectiveness for binary classification using logistic loss yet indeterministic for multi-class classification using softmax loss; (2) disadvantage of logistic loss for single-label multi-class classification via one-vs.-all approach, which is due to the averaging effect on predicted probabilities for the negative class (e.g., non-target classes) in the learning process. With the disadvantage and advantage of logistic loss disentangled, we thereafter propose a novel reweighted logistic loss for multi-class classification. Our simple yet effective formulation improves ordinary logistic loss by focusing on learning hard non-target classes (target vs. non-target class in one-vs.-all) and turned out to be competitive with softmax loss. We evaluate our method on several benchmark datasets to demonstrate its effectiveness.

THE INFLUENCE OF INQUIRY AND DISCOVERY LEARNING METHOD AND CREATIVITY LEVEL IN WRITING SKILLS DESCRIPTION IN NAUTIC TARUNA AND TECHNIQUE IN SURABAYA SHIPPING POLYTECHNIC

2020 ◽  
Vol 8 (1) ◽  
Author(s):  
Ni Putu Dian Permata Prasetyaningrum
Keyword(s):  
Quantitative Methods ◽  
Writing Skills ◽  
Discovery Learning ◽  
Learning Method ◽  
Test Results ◽  
Learning Methods ◽  
Is Research ◽  
Investigation Methods ◽  
High Level ◽  
Discovery Method

Surabaya Shipping Polytechnic emphasizes on certain areas of expertise that Taruna must possess. This is the basis after graduating from shipping polytechnics, cadets must have expertise and skills. The purpose of this study was to study the effect of inquiry, discovery learning, and creativity levels on the ability to write descriptive essays on nautical and technical cadets at Surabaya Shipping Polytechnic. This type of research is research. This research uses quantitative methods using experiments. The location used in this research is Surabaya Shipping Polytechnic. The subjects in this study were the cadets of the Nautika A, Nautika B, Teknika A, and Teknika B. classes. Based on the results of the research and discussion, the following conclusions are obtained: There are those that can be solved looking for description essays in the cadets. learning discovery method. The test results show better investigation methods than the discovery of learning, There is a difference in the ability to write a description essay about cadets who have a high level of creativity with cadets who have a low level of creativity, the test results show better who have a high level of creativity, there are related with learning methods and descriptions of the ability to write essay descriptions, the test results show learning methods and creativity descriptions of the ability to write essay descriptions.

A Study on Multi Class Classification from Breast Cancer Images using Ensemble Network and Transfer Learning

2020 ◽  
Vol 14 ◽  
Author(s):  
Lahari Tipirneni ◽  
Rizwan Patan
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Convolutional Neural Network ◽  
Binary Classification ◽  
Disease Diagnosis ◽  
Feature Descriptors ◽  
Histopathological Images ◽  
Viable Approach ◽  
Multi Class Classification

Abstract:: Millions of deaths all over the world are caused by breast cancer every year. It has become the most common type of cancer in women. Early detection will help in better prognosis and increases the chance of survival. Automating the classification using Computer-Aided Diagnosis (CAD) systems can make the diagnosis less prone to errors. Multi class classification and Binary classification of breast cancer is a challenging problem. Convolutional neural network architectures extract specific feature descriptors from images, which cannot represent different types of breast cancer. This leads to false positives in classification, which is undesirable in disease diagnosis. The current paper presents an ensemble Convolutional neural network for multi class classification and Binary classification of breast cancer. The feature descriptors from each network are combined to produce the final classification. In this paper, histopathological images are taken from publicly available BreakHis dataset and classified between 8 classes. The proposed ensemble model can perform better when compared to the methods proposed in the literature. The results showed that the proposed model could be a viable approach for breast cancer classification.

Sign in / Sign up

Export Citation Format

Share Document