- Ch. 1, p. 5: In Eq. 1.4, the first line should be

I= 0.60R – 0.28G – 0.32B - Ch. 2, p. 24: There are sign errors in the rotation matrix given by (2.10). The correct matrix is (in LateX notation){\bf R} = \left[ \begin{array}{lll}
q_0^2-q_1^2-q_2^2+q_3^2 & 2(q_0q_1-q_2q_3) & 2(q_0q_2+q_1q_3) \\

2(q_0q_1+q_2q_3) & -q_0^2+q_1^2-q_2^2+q_3^2 & 2(q_1q_2-q_0q_3) \\

2(q_0q_2-q_1q_3) & 2(q_1q_2+q_0q_3) & -q_0^2-q_1^2+q_2^2+q_3^2

\end{array} \right]

- Ch. 3, p. 45: in Fig. 3.9(c) the vector u1 should be depicted perp. to v2 and u2 perp. to v1.
- Ch. 3, p. 49: Eqs. 3.36 – 3.38 and 3.42 use (f+k) as opposed to (f-k) used in Eq. 3.22 -3.24 on page 46. This is not an error but an inconsistency in notation.
- Ch. 4, p. 61, Fig. 4.5: “h(k-n)” should be “h(n-k)”
- Ch. 4, p. 67, 1st line, “matrix of integers” should be “matrix of rational numbers.”
- Ch. 5, p. 75, four lines from the bottom: “velocity” and “displacement” should be interchanged.
- Ch. 5, p. 81: the line below Eq. 5.4 should read “x_1 and x_2 vary by…” instead of “x_1 and x_2 varies by…”
- Ch. 5, p. 82, Eq. 5.9: the matrix to be inverted is missing two \partial symbols in the cross diagonal terms. The correct matrix should be\left[ \begin{array}{ll}
\frac{\partial^2 s_c({\bf x},t)}{\partial x_1^2} &

\frac{\partial^2 s_c({\bf x},t)}{\partial x_2 \partial x_1} \\

\frac{\partial^2 s_c({\bf x},t)}{\partial x_1 \partial x_2} &

\frac{\partial^2 s_c({\bf x},t)}{\partial x_2^2}

\end{array} \right]^{-1}

- Ch. 5, p. 90: Equation 5.31 is missing a 1/M normalization factor in the denominator (M = number of pixels in the image).
- Ch. 5, p.94: Page numbers for the reference [Ura 88] should be “79-87” instead of “79-97”
- Ch. 7, pp.122-123: Eqns. 7.16 – 7.21 should use s(x+d,t+dt) instead of s(x-d,t-dt). This is because a Taylor series expansion is considered about (x+d^i,t+dt), and hence the gradient should be computed at (x+d^i,t+dt).
- Ch. 7, p.126, five lines below (7.27): A distinction must be made between the identity matrix in R_u which is a 2×2 identity matrix and the one in R_n which is an NxN identity matrix. They should be denoted by I_2 and I_N, respectively.
- Ch. 8, p.139, “auxilary” should be “auxiliary”
- Ch.14, p.264, Section 14.1.1 Line 6: “LLMSE” should be “LMMSE”
- Ch.14, p.268, In Eqs. 14.24 and 14.25 \mu_s, \mu_g, and \sigma^2_g are not constants but vary from pixel to pixel. That is, \mu_s(n1,n2)= \sum_{(i,j)\in W}g(n1-i,n2-j), etc.
- Ch.16, p.316: Eqns. 16.3 – 16.14 use t_e and t_o for even and odd fields, while 16.15 – 16.16 use t_i and t_{i-1}. This is another inconsistency in the notation.
- Ch.18, p.351, Eqn. 18.3: The notation $H(X)$ does not denote the marginal entropy, it is meant to indicate the conditional entropy of the Kth order Markov source. Furthermore, there is a minus sign missing before the summation in the following mathematical equation.
- Ch.18, p.359: the example in Fig. 18.5 should be corrected as follows\> \underline{Received Bit} \> \underline{Interval} \> \underline{Symbol} \\
\> 1 \> $\left[ 0.5,1 \right)$ \> – \\

\> 0 \> $\left[ 0.5,0.75 \right)$ \> $a_2$ \\

\> 0 \> $\left[ 0.5,0.625 \right)$ \> $a_1$ \\

\> 1 \> $\left[ 0.5625,0.625 \right)$ \> – \\

\> 1 \> $\left[ 0.59375,0.625 \right)$ \> – \\

\> 0 \> $\left[ 0.59375,0.609375 \right)$ \> $a_3$ \\

\> \vdots \> \vdots

- Ch.18, p.361: Eqn. 18.7 and Fig. 18.7 are inconsistent. Eqn. 18.7 should be “(a+c)/2” not “(a+b)/2”
- Ch.20, p.391: Fig. 20.1 “b” should be “b_2” (subscript missing).
- Ch.24, p.480: caption of Fig. 24.11, 2nd line “..the decoded the third..” should be “..the decoded third..”
- Ch.25, p.496, last sentence of the first paragraph: “is” should be “are”
- App.A, p.502, line 3: “such” should be “such as”

PLEASE E-MAIL ANY OTHER ERRORS/CORRECTIONS TO

tekalp@ee.rochester.edu

YOUR HELP IS GREATLY APPRECIATED.