Hasil Pencarian  ::  Simpan CSV :: Kembali

Abstrak :
The purpose of this research is to examine the influence of good corporate governance mechanism about earnings management in companies listed in Indonesian Stock Exchange during 2006 to 2010. The independent variables include the size of commissioner board, independent commissioner board percentage, size of audit committee, and commissioner meeting frequency. The dependent variable is earnings management which is measured by discretionary revenue model (Stubben, 2010). Size of company is used as the control variable in this research. The population of this research is 465 samples from companies listed at Indonesian Stock Exchange during 2006 to 2010. The sampling method used in this research is purposive sampling method. In addition, the data analysis method used is regression analysis and descriptive statistics. The result of this research indicates that the mechanism of good corporate governance which is represented by the size of commissioner board, independent commissioner board percentage, size of audit committee, and commissioner meeting frequency do not have any significant impact on earnings management. However, the result shows that company size gave positive influence toward earning management.

Abstrak :
Pada penelitian ini, dikemukakan sebuah metode baru berbasis analisis multiresolusi untuk mendeteksi distorsi blok pada gambar digital terkompresi. Gambar digital terkompresi cenderung memiliki artefak codingyang mungkin muncul ketika gambar dikodekan dengan tingkat kompresi yang tinggi. Penelitian ini berfokus pada distorsi blok yang dirasakan signifikan dalam gambar digital terkompresi berbasis blok seperti JPEG. Pada penelitian ini, transformasi Wavelet Haar digunakan untuk mendekomposisi sebuah gambar dan menganalisis karakteristik tepian dari gambar tersebut. Berdasarkan dekomposisi ini, peneliti menyusun sebuah algoritma untuk mendeteksi distorsi blok dengan menganalisis koefisien hasil transformasi wavelet. Hasil eksperimen algoritma terhadap database gambar LIVE menunjukkanhasil yang sangat memuaskan dengan tingkat kesalahan yang rendah.

---

Abstract
In this study, presented a new method based on multiresolution analysis to detect the distortion of the block in a compressed digital image. Compressed digital image tend to have coding artifacts that may arise when an image is encoded with a high compression rate. This study focuses on a block distortion that significantly perceived in the block-based compressed digital images such as JPEG. In this study, Wavelet Haar transformation is used to decompose an image and analyze the characteristics of the edge of the picture. Based on this decomposition, the researchers compiled an algorithm for detecting a block distortion by analyzing the coefficients of the wavelet transformation. The results of experimental algorithms for image database LIVE shows very satisfactory results with low error rates.