62
Biomedical Signal and Image Processing
to capture a large number of brain images that need to be saved in physical
medium such as a hard disk or a CD. Therefore, it is essential to compress
these images to avoid the excessive cost of digital storage. Even though the
image in this problem is already compressed (using JPEG technology), we
would like to explore compressing this image using bit slicing.
a. Eliminate the three LSBs of the image and compare the quality of the resulting
image with the original one.
b. Eliminate the four LSBs of the image and compare the quality of the resulting
image with the original one.
c. Continue eliminating the bits as long as the quality of the resulting image is
visually satisfactory. How many bits can be eliminated before the quality
is unsatisfactory? What compression percentage is achieved?
3.3 Read the mouse vertebra image “p_3_3.jpg”.* Using MATLAB, perform
histogram equalization and compare the quality of the resulting equalized
image with the original one.
3.4 Load the MRI image in “p_3_4.mat.” This image is essentially the MRI image of
Problem 3.2 that is corrupted by additive noise. As can be seen, the quality of the
image is rather poor due to the noise. Conduct the following processing steps using
MATLAB to improve the image quality:
a. Use the low-pass masks shown in Figure 3.10 to filter the image.
b. Compare the visual performance of the two masks.
c. Design a similar mask (by changing the numbers in the aforementioned
masks) to outperform the masks used in part “a.”
d. U se a 3 × 3 median filter to filter the image and compare the resulting image
with those of the previous parts.
3.5 L oad the image in “p_3_5.mat.” In this image, we would like to improve the
quality of the image by sharpening of the edges. Conduct the following processing steps using MATLAB:
a. Apply a high-boost filter on the image using A = 1, 1.05, 1.10, 1.15, and 1.20.
b. Co mpare the results of the part “b” and identify the value of A that gives the
best performance.
c. U se a derivative filter to emphasize the edges and compare the results with
those of part “a.”
d. How would you modify the derivative filters for image improvement
applications?
REFERENCE
Goldberger, A.L., Amaral, L.A.N., Glass, L., Hausdorff, J.M., Ivanov, P.Ch., Mark, R.G.,
Mietus, J.E., Moody, G.B., Peng, C.K., and Stanley, H.E. (2000, June 13). PhysioBank,
PhysioToolkit, and PhysioNet: Components of a new research resource for complex
physiologic signals. Circulation 101(23):e215–e220. [Circulation Electronic Pages;
http://circ.ahajournals.org/cgi/content/full/101/23/e215].
* Courtesy of Dr. Helen Gruber, Carolina Medical Center, Charlotte, NC.
Biomedical Signal and Image Processing
to capture a large number of brain images that need to be saved in physical
medium such as a hard disk or a CD. Therefore, it is essential to compress
these images to avoid the excessive cost of digital storage. Even though the
image in this problem is already compressed (using JPEG technology), we
would like to explore compressing this image using bit slicing.
a. Eliminate the three LSBs of the image and compare the quality of the resulting
image with the original one.
b. Eliminate the four LSBs of the image and compare the quality of the resulting
image with the original one.
c. Continue eliminating the bits as long as the quality of the resulting image is
visually satisfactory. How many bits can be eliminated before the quality
is unsatisfactory? What compression percentage is achieved?
3.3 Read the mouse vertebra image “p_3_3.jpg”.* Using MATLAB, perform
histogram equalization and compare the quality of the resulting equalized
image with the original one.
3.4 Load the MRI image in “p_3_4.mat.” This image is essentially the MRI image of
Problem 3.2 that is corrupted by additive noise. As can be seen, the quality of the
image is rather poor due to the noise. Conduct the following processing steps using
MATLAB to improve the image quality:
a. Use the low-pass masks shown in Figure 3.10 to filter the image.
b. Compare the visual performance of the two masks.
c. Design a similar mask (by changing the numbers in the aforementioned
masks) to outperform the masks used in part “a.”
d. U se a 3 × 3 median filter to filter the image and compare the resulting image
with those of the previous parts.
3.5 L oad the image in “p_3_5.mat.” In this image, we would like to improve the
quality of the image by sharpening of the edges. Conduct the following processing steps using MATLAB:
a. Apply a high-boost filter on the image using A = 1, 1.05, 1.10, 1.15, and 1.20.
b. Co mpare the results of the part “b” and identify the value of A that gives the
best performance.
c. U se a derivative filter to emphasize the edges and compare the results with
those of part “a.”
d. How would you modify the derivative filters for image improvement
applications?
REFERENCE
Goldberger, A.L., Amaral, L.A.N., Glass, L., Hausdorff, J.M., Ivanov, P.Ch., Mark, R.G.,
Mietus, J.E., Moody, G.B., Peng, C.K., and Stanley, H.E. (2000, June 13). PhysioBank,
PhysioToolkit, and PhysioNet: Components of a new research resource for complex
physiologic signals. Circulation 101(23):e215–e220. [Circulation Electronic Pages;
http://circ.ahajournals.org/cgi/content/full/101/23/e215].
* Courtesy of Dr. Helen Gruber, Carolina Medical Center, Charlotte, NC.