Method for enhancing the contrast of an image

1. A method for enhancing the contrast of an image that is represented by a digital signal representation, comprising the steps of:

decomposing said digital signal representation into a multi-scale representation comprising at least two detail images representing detail at multiple scales;

modifying pixel values of said detail images by at least one of said multiple scales to yield pixel values of a set of modified detail images by applying at least one conversion function to the detail images; and

computing a processed image representation by applying a reconstruction algorithm to the modified detail images;

wherein said conversion function is a monotonically increasing odd conversion function providing maximum amplification within an intermediate subrange of pixel values which intermediate subrange is considered most critical, said conversion function represented by a curve having a steep amplification peak causing a slope of the curve to evolve to a local minimum following a local maximum wherein the slope of the curve increases after the local minimum with the curve being concave in a positive y(x) direction after the amplification peak.

2. A method according to claim 1 wherein layers of said multi-scale representation are pixel-wise multiplied with a coefficient that depends upon said at least one scale.

3. A method according to claim 2 wherein said coefficient is greater than one at a finest of said multiple scales and gradually decreases to a value equal to one at a medium scale of said multiple scales and remains one at a larger of said multiple scales.

4. A method according to claim 1 wherein enhancement of critical details is maximal at scale zero and decreases according to a geometric series at medium scales and is constant at larger scales.

5. A method according to claim 1 wherein a point of maximal contrast boosting is made proportional to a noise level.

6. A method according to claim 1 wherein said conversion function is derived from a parameterised conversion function q(x) by giving specific values to parameters of said parameterised conversion function, wherein specific values for examination type dependent parameters are exclusively derived from said digital signal representation.

7. A method according to claim 1 , wherein each of the detail images comprises differences of gray values of approximation images of the digital signal representation at different scales.

8. A method for enhancing the contrast of an image that is represented by a digital signal representation, comprising the steps of:

decomposing said digital signal representation into a multi-scale representation comprising at least two detail images representing detail at multiple scales;

modifying pixel values of said detail images by at least one of said multiple scales to yield pixel values of a set of modified detail images by applying at least one conversion function to the detail images; and

computing a processed image representation by applying a reconstruction algorithm to the modified detail images;

wherein said conversion function is a monotonically increasing odd conversion function providing maximum amplification within an intermediate subrange of pixel values which intermediate subrange is considered most critical, said conversion function represented by a curve having a steep amplification peak causing a slope of the curve to evolve to a local minimum following a local maximum, and

wherein:

layers of said multi-scale representation are pixel-wise multiplied with a coefficient that depends upon the scale;

a segmentation map is constructed wherein with each pixel of said image a label is associated indicating a tissue type which said pixel represents;

a tissue coefficient table is constructed relating a tissue coefficient with a scale and a tissue type;

for each pixel of a layer of multi-scale representation a tissue coefficient corresponding with the scale of said layer and the tissue type of said pixel is retrieved; and

layers of said multi-scale representation are pixel-wise multiplied by corresponding retrieved tissue coefficients.

9. A method according to claim 8 wherein for all scales of the multi-scale representation a maximum amplification is amplified or attenuated by said tissue specific coefficient.

10. A computer program stored on a computer readable carrier medium adapted to carry out the steps of the method of claim 8 .