Image Feature Extraction Using Symbolic Data of Cumulative Distribution Functions DOI Creative Commons
Sri Winarni, Sapto Wahyu Indratno,

Restu Arisanti

и другие.

Mathematics, Год журнала: 2024, Номер 12(13), С. 2089 - 2089

Опубликована: Июль 3, 2024

Symbolic data analysis is an emerging field in statistics with great potential to become a standard inferential technique. This research introduces new approach image feature extraction using the empirical cumulative distribution function (ECDF) and of values (DFDV) as symbolic data. The main objective reduce dimension huge pixel by organizing them into more coherent pixel-intensity distributions. We propose partitioning method different breakpoints capture intensity variations effectively. results ECDF representing proportion intensities DFDV probability at specific points. novelty this lies features, thus summarizing providing informative representation value distribution, facilitating classification based on distribution. experimental underscore distinguishing characteristics among existing classes. Image features extracted promise representations. In addition, theoretical insights properties functions are gained.

Язык: Английский

Redesigning urban parks for active living in dense urban areas: a remote audit approach DOI
Yufeng Luo, Andrew T. Kaczynski, Jiuling Li

и другие.

International Journal of Environmental Health Research, Год журнала: 2025, Номер unknown, С. 1 - 11

Опубликована: Март 3, 2025

The widespread use of virtual audits has transformed the assessment urban design attributes by eliminating necessity for on-site visits. However, there remains a lack audit tools specifically designed evaluating parks in dense environments. This study aims to (1) adapt Audit Tool Activity-friendly Parks Dense Urban Areas (TAPS) into remote tool (R-TAPS), and (2) evaluate its reliability validity. Trained auditors used R-TAPS conduct (n = 53) Tokyo, with subset 25). Kappa statistics percent agreement assessed inter-rater reliability, intra-class correlation coefficient (ICC) verified convergent showed moderate almost perfect 89% items. Remote exhibited high positive (ICC 0.73). offers reliable virtually promote physical activity settings.

Язык: Английский

Процитировано

0
Image Feature Extraction Using Symbolic Data of Cumulative Distribution Functions DOI Creative Commons
Sri Winarni, Sapto Wahyu Indratno,

Restu Arisanti

и другие.

Mathematics, Год журнала: 2024, Номер 12(13), С. 2089 - 2089

Опубликована: Июль 3, 2024

Symbolic data analysis is an emerging field in statistics with great potential to become a standard inferential technique. This research introduces new approach image feature extraction using the empirical cumulative distribution function (ECDF) and of values (DFDV) as symbolic data. The main objective reduce dimension huge pixel by organizing them into more coherent pixel-intensity distributions. We propose partitioning method different breakpoints capture intensity variations effectively. results ECDF representing proportion intensities DFDV probability at specific points. novelty this lies features, thus summarizing providing informative representation value distribution, facilitating classification based on distribution. experimental underscore distinguishing characteristics among existing classes. Image features extracted promise representations. In addition, theoretical insights properties functions are gained.

Язык: Английский

Процитировано

2