Features Extraction Based on Linear Regression Technique

Abstract

Problem statement:The matching problem of complex objects is one of the most difficult task in the pattern recognition field. These problems are made difficult by seemingly infinite varieties of shapes and classes which are used. The difficulties are related to absolute shape measurement, given the impossibility of directly mapping shapes, as such, into a feature space. Approach: In this study, an object was modeled using boundaries pixel distance. The invariant has been resulted from the distance of each boundaries pixel to their central point. By performing linear regression on each set of sorted distances, a unique set of numerical features from the coefficients of this linear function has been produced. This unique set of numerical values is then proposed as an object's features. Results: The experiments show that the coefficient of linear function from boundaries' distance plot of each object has produced better recognition than polynomial function of degree more than one. Conclusion/Recommendations: More than 200 hundreds trademark's images have been tested and almost 90% of successful rate of accuracy has been achieved.

Scalable prediction-based online anomaly detection for smart meter data

“…Zhang et al further considered the impact of outdoor temperature and used a three-piecewise linear regression method to fit the relationship between energy consumption and outdoor temperature. However, linear regression-based methods require well-defined independent variables [31]. Brown et al predicted electricity demand using K-nearest neighborhood (KNN) in a kernel regression method [5].…”

“…Clustering-based 5,6,7,9,11,12,13,14,15,16,17,18,21,30,31 Boxplot 5,6,7,9,10,11,12,13,14,15,16,17,18,19,21,30,31 Prediction-based (PARX) 5,6…”

Probabilistic anomaly detection in natural gas time series data

“…Therefore, penalty function strategies do not always guarantee practical results. The advantage of linear regression is that, with the dependent variables being well defined, the technique is able to extract time series features (Magld, 2012). Lee and Fung (1997) showed that linear and nonlinear regressions can also be used for outlier detection, but they used a 5% upper and lower threshold limit for choosing outliers after fitting, which yielded many false positives for very large data sets.…”

Regression-based Online Anomaly Detection for Smart Grid Data

“…Jakkula and Cook use statistics and clustering to identify outliers in power datasets collected from smart environments [14], but they have not considered the impact of the exogenous variables, e.g., weather temperature, on the electricity consumption. Linear regression can extract time series features when the dependent variables are well-defined [31]. The early experience of identifying outliers in linear regression is through setting a threshold limit, but this yields many false positives for large data sets [18].…”