A generalization of Otsu method for linear separation of two unbalanced classes in document image binarization

The classical Otsu method is a common tool in document image binarization. Often, two classes, text and background, are imbalanced, which means that the assumption of the classical Otsu method is not met. In this work, we considered the imbalanced pixel classes of background and text: weights of two classes are different, but variances are the same. We experimentally demonstrated that the employment of a criterion that takes into account the imbalance of the classes' weights, allows attaining higher binarization accuracy. We described the generalization of the criteria for a two-parametric model, for which an algorithm for the optimal linear separation search via fast linear clustering was proposed. We also demonstrated that the two-parametric model with the proposed separation allows increasing the image binarization accuracy for the documents with a complex background or spots.

Linear Method for Diagnosis of Inter-Turn Short Circuits in 3-Phase Induction Motors

When a turn-to-turn short fault occurs in an induction motor, it will be accompanied by vibration and heating, which will have adverse effects on the entire power system. Thus, turn-to-turn short fault diagnosis of the stator is required, and major accidents can be prevented if an inter-turn short circuit (ITSC), which is the early stage of a turn-to-turn short, can be detected. This study reinterprets Park’s vector approach using Direct-Quadrature(D-Q) transformation for the linear separation of ITSCs and proposes an ITSC diagnosis method by defining the magnetic flux linkage pulsation and current change in the event of a turn-to-turn short. It is difficult to diagnose because the turn-to-turn short current change in an ITSC is considerably different from the induction motor loss. Hence, it was found through analysis that when the current change is considered through an analysis of the relationship between inductance and the winding number, the ITSC current becomes slightly smaller than the steady-state current. This was verified using the D-Q synchronous reference frame over time. We proposed a linear separation of the ITSC diagnosis from the steady state by considering the minimum values of the pulsating current as feature points.

Orthogonal linear separation analysis: an approach to decompose the complex effects of a perturbagen

Drugs have multiple, not single, effects. Decomposition of drug effects into basic components helps us to understand the pharmacological properties of a drug and contributes to drug discovery. We have extended factor analysis and developed a novel profile data analysis method, orthogonal linear separation analysis (OLSA). OLSA contracted 11,911 genes to 118 factors from transcriptome data of MCF7 cells treated with 318 compounds in Connectivity Map. Ontology of the main genes constituting the factors detected significant enrichment of the ontology in 65 of 118 factors and similar results were obtained in two other data sets. One factor discriminated two Hsp90 inhibitors, geldanamycin and radicicol, while clustering analysis could not. Doxorubicin was estimated to inhibit Na+/K+ATPase, one of the suggested mechanisms of doxorubicin-induced cardiotoxicity. Based on the factor including PI3K/AKT/mTORC1 inhibition activity, 5 compounds were predicted to be novel autophagy inducers, and other analysis including western blotting revealed that 4 of the 5 actually induced autophagy. These findings indicate the potential of OLSA to decompose the effects of a drug and identify its basic components. (<175 words)