Image completion with multi-image based on entropy reduction

Wu, Hao; Zhang, Miao; Wang, Yi; Chen, Jingyue; Ma, Cong; Zhou, Tianyu

doi:10.1016/j.neucom.2014.12.088

Cited by 33 publications

(10 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is probably because of two factors: 1) our proposed scheme is less sensitive to initial labeled data than supervised learning based schemes, as it has the ability to use the information from the unlabeled data to improve its performance 2) the limitation of insufficient labeled data will become more obvious if the size of labeled data is extremely small (like 30 in our study), the performance improvement of using unlabeled data would become limited. Since our algorithm requires splitting the initial labeled dataset, if the size is extremely small the split dataset will be even smaller which makes the scheme more sensitive to noise [31][32].…”

Section: Discussionmentioning

confidence: 99%

Computerized breast cancer analysis system using three stage semi-supervised learning method

Sun

Tseng

Zhang

et al. 2016

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Computerized breast cancer analysis system using three stage semi-supervised learning method

Sun

Tseng

Zhang

et al. 2016

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

“…To apply the entropy method, firstly, the image is divided into small blocks and the entropy En is calculated for each block as follows [18,19]:…”

Section: Proposed Algorithmmentioning

confidence: 99%

Clouds Motion Estimation from Ground-Based Sky Camera and Satellite Images

Zaher¹,

Ghanem²

2018

Colorimetry and Image Processing

View full text Add to dashboard Cite

Estimation of cloud motion is a challenging task due to the non-linear phenomena of cloud formation and deformation. Satellite images processing is a popular tool used to study the characteristics of clouds which constitute major factors in forecasting the meteorological parameters. Due to the low resolution of satellite images, researchers have turned towards analyzing the high-resolution images captured by ground-based sky cameras. The first objective of this chapter is to demonstrate the different techniques used to estimate clouds motion and to compare them with respect to the accuracy and the computational time. The second aim is to propose a fast and efficient block matching technique based on combining the two types of images. The first idea of our approach is to analyze the low-resolution satellite images to detect the direction of motion. Then, the direction is used to orient the search process to estimate the optimal motion vectors from the high-resolution ground-based sky images. The second idea of our method is to use the entropy technique to find the optimal block sizes. The third idea is to imply an adaptive cost function to perform the matching process. The comparative study demonstrates the high performance of the proposed method with regards to the robustness, the accuracy and the computation time.

show abstract

“…By directly comparing the testing motion data with a few prototype template generated from different classes called motifs, we are able to tell the type of motion the testing data belongs to. The main idea of the previous related researches focus on finding the optimized samples that can better represent the features of the categories they belong to [26][27][28][29][30][31]. Some studies also consider the separability between different classes [32].…”

Section: Related Workmentioning

confidence: 99%

Sports Motion Recognition Using MCMR Features Based on Interclass Symbolic Distance

Wei

Jiao

Wang

et al. 2016

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

Human motion and gesture recognition receive much concern in sports field, such as physical education and fitness for all. Although plenty of mature applications appear in sports training using photography, video camera, or professional sensing devices, they are either expensive or inconvenient to carry. MEMS devices would be a wise choice for students and ordinary body builders as they are portable and have many built-in sensors. In fact, recognition of hand gestures is discussed in many studies using inertial sensors based on similarity matching. However, this kind of solution is not accurate enough for human movement recognition and cost much time. In this paper, we discuss motion recognition in sports training using features extracted from distance estimation of different kinds of sensors. To deal with the multivariate motion sequence, we propose a solution that applies Max-Correlation and Min-Redundancy strategy to select features extracted with interclass distance similarity estimation. With this method, we are able to screen out proper features that can distinguish motions in different classes effectively. According to the results of experiment in real world application in dance practice, our solution is quite effective with fair accuracy and low time cost.

show abstract

Image completion with multi-image based on entropy reduction

Cited by 33 publications

References 36 publications

Computerized breast cancer analysis system using three stage semi-supervised learning method

Computerized breast cancer analysis system using three stage semi-supervised learning method

Clouds Motion Estimation from Ground-Based Sky Camera and Satellite Images

Sports Motion Recognition Using MCMR Features Based on Interclass Symbolic Distance

Contact Info

Product

Resources

About