In statistical image reconstruction, data are often recorded on a regu
lar grid of squares, known as pixels, and the reconstructed image is d
efined on the same pixel grid. Thus, the reconstruction of a continuou
s planar image is piecewise constant on pixels, and boundaries in the
image consist of horizontal and vertical edges lying between pixels. T
his approximation to the true boundary can result in a loss of informa
tion that may be quite noticeable for small objects, only a few pixels
in size. increasing the resolution of the sensor may not be a practic
al alternative. If some prior assumptions are made about the true imag
e, however, reconstruction to a greater accuracy than that of the reco
rding sensor's pixel grid is possible. We adopt a Bayesian approach, i
ncorporating prior information about the true image in a stochastic mo
del that attaches higher probability to images with shorter total edge
length. in reconstructions, pixels may be of a single color or split
between two colors. The model is illustrated using both real and simul
ated data.