Abstract
This research emphasizes on problem of determining the suitable threshold value in image compression without harming the image quality that may affect the Human Visual System (HVS). A technique is proposed by using wavelet-based subband thresholding technique and standard deviation value is applied to estimate the near zero valued coefficients. By using the standard deviation value as threshold value, the near zero coefficients is suppressed to zero to produce a compressed image. The results are then compared to the previous compression algorithms. From the analysis, it shows that although some coefficients were discarded using the thresholding process, the quality of the image is still maintained at acceptable level.