If you are building a presentation based on S. Jayaraman's material, keep these slide design principles in mind:
: Based on derivatives. The Laplacian operator uses the second derivative for high-frequency edge accentuation. 4. Image Enhancement in the Frequency Domain
Information ignored by the human visual system.
Region Growing, Region Splitting, and Merging. digital image processing jayaraman ppt
Segmentation algorithms are generally based on one of two basic properties of intensity values: discontinuity and similarity. Detection of Discontinuities (Edge-Based) Finding abrupt changes in intensity.
: Mathematical foundations including 2D convolution, Z-transforms, and popular image transforms like Fourier or Discrete Cosine Transform (DCT) .
Human eye perceives color via primary colors (Red, Green, Blue) and secondary colors (Magenta, Cyan, Yellow). Color Models: RGB Model: Used for computer monitors and cameras. CMY/CMYK Model: Used for color printing. If you are building a presentation based on S
Improving quality (e.g., contrast enhancement).
Crucial for calibrating contrast display devices (Gamma correction). s=c⋅rγs equals c center dot r raised to the gamma power : Expands dark values (similar to log transform). : Compresses dark values and expands bright values. 2.2 Histogram Processing
F̂(u,v)=[1H(u,v)|H(u,v)|2|H(u,v)|2+Sη(u,v)Sf(u,v)]G(u,v)cap F hat open paren u comma v close paren equals open bracket the fraction with numerator 1 and denominator cap H open paren u comma v close paren end-fraction the fraction with numerator the absolute value of cap H open paren u comma v close paren end-absolute-value squared and denominator the absolute value of cap H open paren u comma v close paren end-absolute-value squared plus the fraction with numerator cap S sub eta open paren u comma v close paren and denominator cap S sub f open paren u comma v close paren end-fraction end-fraction close bracket cap G open paren u comma v close paren Sηcap S sub eta Sfcap S sub f are the power spectra of the noise and the original image. Module 5: Image Segmentation and Compression 5.1 Image Segmentation Segmentation algorithms are generally based on one of
Summary, Conclusion, and Reference to S. Jayaraman’s Text
Uses first-order derivatives (Gradient operators like Sobel, Prewitt, and Robert's cross) and second-order derivatives (Laplacian, Laplacian of Gaussian).
Directly dividing the degraded image transform by the degradation function. It fails in the presence of noise.