WO2006082542A1

WO2006082542A1 - Clipping

Info

Publication number: WO2006082542A1
Application number: PCT/IB2006/050268
Authority: WO
Inventors: Erno H. A. Langendijk
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2005-02-07
Filing date: 2006-01-25
Publication date: 2006-08-10

Abstract

An image conversion unit (100) for converting an input image (110) into an output image (116) is disclosed. The input image (110) comprises input pixels having respective values in an input range of values (200) and the output image (116) comprises output pixels having respective values in an output range of values (202), wherein the output range of values (202) is smaller than the input range of values (200). The image conversion unit (100) comprises: a filter (102) for establishing that a first one (114,232) of the input pixels of the input image (110) has a first value which exceeds the output range of values (202); and a pixel processor (104) for assigning a second value to a second one (118,236) of the output pixels on basis of a third value of a second one (112,230) of the input pixels and on basis of a spatial distance between the first one (114,232) of the input pixels and the second one (112,230) of the input pixels, the second one (112,230) of the input pixels and the second one (118,236) of the output pixels having mutually equal spatial coordinates.

Description

Clipping

The invention relates to an image conversion unit for converting an input image into an output image, the output image comprising output pixels having respective values in an output range of values.

The invention further relates to an image processing apparatus, comprising: - receiving means for receiving an input image; and an image conversion unit for converting the input image into an output image, the output image comprising output pixels having respective values in an output range of values.

The invention further relates to a method of converting an input image into an output image, the output image comprising output pixels having respective values in an output range of values.

The invention further relates to a computer program product to be loaded by a computer arrangement, comprising instructions to convert an input image into an output image, the output image comprising output pixels having respective values in an output range of values, the computer arrangement comprising processing means and a memory.

Clipping is a very common operation in image processing and video processing. Clipping is often performed when the intermediate values of a particular processing unit exceed the range of output values. Values below or above the range of output values are typically set to the minimum and maximum of the range of output values, respectively. Processing operations, which often result in a relatively large range of values, i.e. typically exceeding the range of output values, are luminance enhancement and color enhancement operations. Typically, in video broadcasting standards, such as the EBU, video signals for red, green and blue (RGB) should be within 16 to 235. The range from 0 to 16 and from 235 to 255 is used for under and overshoots for video processing. Yet, after processing the values below 16 and above 235 have to be clipped.

A problem of clipping is that clearly visible artefacts may be introduced. It is an object of the invention to provide an image conversion unit of the kind described in the opening paragraph, to substantially prevent the introduction of clearly visible clipping artifacts.

This object of the invention is achieved in that the image conversion unit comprises: a filter for establishing that a first one of the input pixels of the input image has a first value which exceeds the output range of values; and - a pixel processor for assigning a second value to a second one of the output pixels on basis of a third value of a second one of the input pixels and on basis of a spatial distance between the first one of the input pixels and the second one of the input pixels, the second one of the input pixels and the second one of the output pixels having mutually equal spatial coordinates. Clipping artifacts are typically introduced because the values of a group of connected input pixels are all mapped to the same value of the output range of values: typically, the maximum or minimum value of the output range of values. Then the mapping is a unitary operation, i.e. only based on the input pixel value itself. The image conversion unit according to the invention is arranged to adapt not only a particular pixel value which exceeds the range of output values but is also arranged to adapt further pixel values of further pixels which are located in the neighborhood of the particular pixel. So, for assigning the second value to the second one of the pixels, the image conversion unit according to the invention is arranged to establish that the first one of the input pixels exceeds the output range of values and on base of that the pixel processor performs the assignment of the second value, wherein the actually assigned second value is related to the spatial distance between the first one of the input pixels and the second one of the input pixels. Preferably, the relation between the difference between the second value and the third value, and the spatial distance is inverse proportional. That means that the difference between the second value and the third value is relatively low if the spatial distance is relatively large. As a result, the various values of the group of connected input pixels are adapted differently.

The values may be related to luminance and/or color.

In an embodiment of the image conversion unit according to the invention, the pixel processor is arranged to assign the second value to the second one of the output pixels, wherein the second value is lower than the third value of the second one of input pixels, if the first value of the first one of the input pixels is higher than the maximum value of the output range of values. In other words, relatively high values of the input image are decreased to fit into the range of output values.

In an embodiment of the image conversion unit according to the invention, the pixel processor is arranged to assign the second value to the second one of the output pixels, wherein the second value is higher than the third value of the second one of input pixels, if the first value of the first one of the input pixels is lower than the minimum value of the output range of values. In other words, relatively low values of the input image are increased to fit into the range of output values. In an embodiment of the image conversion unit according to the invention, the filter has an aperture for establishing that the first one of the input pixels of the input image has a first value which exceeds the output range of values, which simultaneously covers the first one of the input pixels and the second one of the input pixels and which is substantially smaller than the size of the input image. For assigning the second value to the second one of the output pixels, the values of the pixels in the neighborhood of the second one of the input pixels have to be checked. Typically this checking is performed by means of a sliding window with a predetermined size, related to the aperture. An appropriate size of the window is e.g. 20*20 pixels for an input image with a size of approximately 500*500 pixels.

In an embodiment of the image conversion unit according to the invention, the pixel processor is arranged to compute the second value of the second one of the output pixels on basis of a first difference between the first value of the first one of the input pixels and the maximum value of the output range of values or the minimum value of the output range of values. That means that for the computation of the second value an exceed level is computed. The exceed level corresponds to the difference between the first value and one of the borders of the output range of values. The exceed level is computed on basis of the minimum of the output range of value if the first value is lower than the minimum. The exceed level is computed on basis of the maximum of the output range of values if the first value is higher than the maximum. Preferably, a relatively high exceed level corresponds to a relatively high difference between the second value of the second one of the output pixels and the third value of the second one of the input pixels. In general, a relatively high exceed level corresponds to a relatively high increase/decrease of corresponding values.

In an embodiment of the image conversion unit according to the invention, the pixel processor is arranged to compute the second value of the second one of the output pixels on basis of a second difference between the first value of the first one of the input pixels and the third value of the second one of the input pixels. This embodiment of the image conversion unit according to the invention is also arranged to take into account the actual first value of the first one of input pixels to determine the amount of increase/decrease. Preferably, a relatively high second difference corresponds to a relatively low difference between the second value of the second one of the output pixels and the third value of the second one of the input pixels. In general, a relatively high second difference corresponds to a relatively low increase/decrease of values.

In an embodiment of the image conversion unit according to the invention, the pixel processor is arranged to compute the second value of the second one of the output pixels by means of multiplying the third value of the second one of the input pixels with a factor which is based on the spatial distance between the first one of the input pixels and the second one of the input pixels. Preferably, the factor is determined by a monotone function of spatial distance and by the spatial distance between the first one of the input pixels and the second one of the input pixels. The function may correspond to any function of the set of functions comprising: cosine, square cosine and Gaussian.

It is advantageous that the factor is based on a first difference between the first value of the first one of the input pixels and the maximum value of the output range of values or the minimum value of the output range of values. Preferably, the factor is proportional to the first difference. It is advantageous that the factor is based on a second difference between the first value of the first one of the input pixels and the third value of the second one of the input pixels. Preferably, the factor is inversely proportional to the second difference.

In an alternative embodiment of the image conversion unit according to the invention, the pixel processor is arranged to compute the second value of the second one of the output pixels by means of subtracting a term from the third value of the second one of the input pixels, wherein the term is based on the spatial distance between the first one of the input pixels and the second one of the input pixels. Preferably, the term is determined by a monotone function of spatial distance and by the spatial distance between the first one of the input pixels and the second one of the input pixels. The iunction may correspond to any function of the set of functions comprising: cosine, square cosine and Gaussian.

It is advantageous that the term is based on a first difference between the first value of the first one of the input pixels and the maximum value of the output range of values or the minimum value of the output range of values. Preferably, the term is proportional to the first difference. It is advantageous that the term is based on a second difference between the first value of the first one of the input pixels and the third value of the second one of the input pixels. Preferably, the term is inversely proportional to the second difference.

It is a further object of the invention to provide an image processing apparatus of the kind described in the opening paragraph, comprising the image conversion unit to substantially prevent the introduction of clearly visible clipping artifacts.

This object of the invention is achieved in that the image conversion unit comprises: a filter for establishing that a first one of the input pixels of the input image has a first value which exceeds the output range of values; and a pixel processor for assigning a second value to a second one of the output pixels on basis of a third value of a second one of the input pixels and on basis of a spatial distance between the first one of the input pixels and the second one of the input pixels, the second one of the input pixels and the second one of the output pixels having mutually equal spatial coordinates.

The image processing apparatus might support one or more of the following types of image processing:

Video compression, i.e. encoding or decoding, e.g. according to the MPEG standard. - De-interlacing: Interlacing is the common video broadcast procedure for transmitting the odd or even numbered image lines alternately. De-interlacing attempts to restore the full vertical resolution, i.e. make odd and even lines available simultaneously for each image;

Image rate conversion: From a series of original input images a larger series of output images is calculated. Output images are temporally located between two original input images; and

Temporal noise reduction. This can also involve spatial processing, resulting in spatial-temporal noise reduction.

The image processing apparatus might e.g. be a TV, a set top box, a VCR (Video Cassette Recorder) player, a satellite tuner, a DVD (Digital Versatile Disk) player or recorder or a Hard-disk recorder.

The image processing apparatus may comprise additional components, e.g. a display device for displaying the output image. It is a further object of the invention to provide a method of the kind described in the opening paragraph, to substantially prevent the introduction of clearly visible clipping artifacts.

This object of the invention is achieved in that the method comprises: - establishing that a first one of the input pixels of the input image has a first value which exceeds the output range of values; and assigning a second value to a second one of the output pixels on basis of a third value of a second one of the input pixels and on basis of a spatial distance between the first one of the input pixels and the second one of the input pixels, the second one of the input pixels and the second one of the output pixels having mutually equal spatial coordinates.

It is a further object of the invention to provide a computer program product of the kind described in the opening paragraph, to substantially prevent the introduction of clearly visible clipping artifacts.

This object of the invention is achieved in that the computer program product, after being loaded, provides said processing means with the capability to carry out: establishing that a first one of the input pixels of the input image has a first value which exceeds the output range of values; and assigning a second value to a second one of the output pixels on basis of a third value of a second one of the input pixels and on basis of a spatial distance between the first one of the input pixels and the second one of the input pixels, the second one of the input pixels and the second one of the output pixels having mutually equal spatial coordinates.

Modifications of the image conversion unit and variations thereof may correspond to modifications and variations thereof of the image processing apparatus, the method and the computer program product, being described.

These and other aspects of the image conversion unit, of the image processing apparatus, of the method and of the computer program product, according to the invention will become apparent from and will be elucidated with respect to the implementations and embodiments described hereinafter and with reference to the accompanying drawings, wherein:

Fig. 1 schematically shows an embodiment of the image conversion unit according to the invention; Fig. 2A schematically shows a mapping of an input value to an output value, wherein the output value is lower than the input value;

Fig. 2B schematically shows a mapping of an input value to an output value, wherein the output value is higher than the input value; Fig. 3 A schematically shows a number of parameters on which the mapping of

Fig. 2A may be based;

Fig. 3B schematically shows a number of parameters on which the mapping of Fig. 2B may be based;

Fig. 4 schematically shows two functions for determining a gain factor to be applied for the mapping of an input value to an output value; and

Fig. 5 schematically shows an embodiment of the image processing apparatus according to the invention.

Same reference numerals are used to denote similar parts throughout the Figures.

Fig. 1 schematically shows an embodiment of the image conversion unit 100, according to the invention. The image conversion unit 100 is arranged to convert input images 110 into output images 116. The input images 110 are provided at the input connector 106 and the output images 116 are outputted at the output connector 108. The input images 110 comprise input pixels 112, 114 having respective values in an input range of values 200 and the output images 116 comprise output pixels 118 having respective values in an output range of values 202, wherein the output range of values 202 is smaller than the input range of values 200. The maximum value 210 of the output range of values 202 corresponds with a value indicated with reference number 212 which is lower than the maximum value 206 of the input range of values 200. The minimum value 208 of the output range of values 202 corresponds with a value indicated with reference number 214 which is higher than the minimum value 206 of the input range of values 200.

The image conversion unit 100 comprises: - a filter 102 for establishing that a first one 114 of the input pixels of the input image 110 has a first value Vl which exceeds the output range of values 202; and a pixel processor 104 for assigning a second value V2 to a second one 118 of the output pixels on basis of a third value V3 of a second one 112 of the input pixels and on basis of a spatial distance S between the first one 114 of the input pixels and the second one 112 of the input pixels, the second one 112 of the input pixels and the second one 118 of the output pixels having mutually equal spatial coordinates.

The filter 102 and the pixel processor 104 may be implemented using one processor. Normally, these functions are performed under control of a software program product. During execution, normally the software program product is loaded into a memory, like a RAM, and executed from there. The program may be loaded from a background memory, like a ROM, hard disk, or magnetical and/or optical storage, or may be loaded via a network like Internet. Optionally an application specific integrated circuit provides the disclosed functionality. The working of the image conversion unit 100 is as follows. The filter 102 comprises a kernel 120, i.e. a sliding window which is arranged to scan over the input image 110. The scan may be based on a row-by-row or column-by-column scheme, wherein the scanning direction is left to right or vice versa, top to bottom or vice versa or zigzag. When processing a second one 112 of the pixels, the pixel values of the pixels in the neighborhood of the second one 112 of the pixels are checked. Checking means that the values of the pixels in the neighborhood are compared with one or optionally two predetermined thresholds. These predetermined thresholds correspond with the borders 208, 210 of the output range of values 202. The first one of these borders corresponds with the minimum value 208 of the output range of values 202. The second one of these borders corresponds with the maximum value 210 of the output range of values 202. The actual neighborhood of the second one 112 of the pixels, which is taken into account, is determined by the size, shape and position of the kernel 120 of the filter 102. The kernel 120 is preferably located with its center aligned with the second one 112 of the pixels. Alternative positions are possible, e.g. the particular pixel is located at one of the borders or corners of the kernel 120. If multiple pixels in the neighborhood exceed the range of values, then it is preferred to select the pixel of those pixels that differs most from the predetermined thresholds. That means that the pixel in the neighborhood having the highest/lowest value is selected. Subsequently, the spatial distance S between the second one 112 of the pixels and the from the neighborhood selected pixel, i.e. the first one 114 of the pixels, is determined. On basis of the actual spatial distance S a gain factor F is determined by means of a function f_x(s) of spatial distance s .

Preferably, the function f_λ(s) corresponds to a cosine:

γ,λ are constants and S_1113x is the maximum distance between two pixels.

Finally, the value V2 of the output pixel 118 having the same coordinates as the second one 112 of the pixels is computed by multiplying the value V3 of the second one 112 of the pixels with the gain factor F , as specified in Equation 3. v_out(χ,y) = v_m(χ,_y)*F (3)

Alternatively, on basis of the actual spatial distance S a term T is determined by means of a function t_x{s) of spatial distance s .

T = φ) (4) Finally, the value V2 of the output pixel 118 having the same coordinates as the second one 112 of the pixels is computed by subtracting the term T from the value V3 of the second one 112 of the pixels, as specified Equation 4.

V_out(x,y) = V_m(x,y) -T (5)

In connection with Figs. 2A,2B,3A,3B and 4 alternative embodiments of the image conversion unit 100 according to the invention are described.

Fig. 2A schematically shows a mapping of an input value V3 to an output value V2 wherein the output value V2 is lower than the input value V3. Fig. 3 A schematically shows a number of parameters S₅E₅D on which the mapping of Fig. 2 A may be based. All pixels of the input image have values which are in the input range of values 200, having a minimum value 204 and a maximum value 206. However, the output range of values 202 has a lower number of values. The maximum value 210 of the output range of values 202 is lower than the maximum value 206 of the input range of values 200. The minimum value 208 of the output range of values 202 is higher than the minimum value 204 of the input range of values 200.

A first one 114 of the pixels of the input image has a value Vl which is higher than the maximum value 210 of the output range of values 202. The image conversion unit 100 according to the invention is arranged to compute the corresponding output value V4. Because the first one 114 of pixels of the input image corresponds to the highest value of pixel values of the input image 110 the corresponding output value V4 is preferably equal to the maximum value 210 of the output range of values 202. Because, Vl is higher than the maximum value 210 of the output range of values 202 and because the spatial distance S between the first one 114 of the pixels of the input image and a second one 112 of the pixels of the input image, is below a predetermined spatial threshold, the value V2 of the output pixel corresponding to the second one 112 of the pixels of the input image is lower than the value V3 of the second one 112 of the pixels of the input image. Preferably, the value V2 of the output pixel 118 corresponding to the second one 112 of the pixels of the input image is computed by multiplying the value V3 of the second one 112 of the pixels of the input image with the gain factor F , as specified in Equation 3.

Preferably, the gain factor F is also dependent on the actual exceed level E, i.e. the difference between the value Vl and the maximum value 210 of the output range of values 202. On basis of the actual spatial distance S and the actual exceed level E a gain factor F is determined by means of a iunction f₂(s,e) of spatial distance s and exceed level e :

F = f₂(s,e) (6) Preferably, the gain factor F is proportional to the exceed level e , e.g.

/₂(*,_e) = λ + -^cos²(-^ *^) (7) ^{α ώ}max ^λ with α a constant.

Preferably, the gain factor F is also dependent on the actual difference D between the value Vl and the value V3. On basis of the actual spatial distance S, the actual exceed level E and the actual difference D a gain factor F is determined by means of a function f₃(s,e,d) of spatial distance s , exceed level eand difference d :

F = f₃(s,e,d) (8)

Preferably, the gain factor F is inversely proportional to the difference d , e.g.: _/3(^_{) = λ +} ^_COs²(-^ *^) (9)

with β a constant.

Fig. 2B schematically shows a mapping of an input value V7 to an output value V6 wherein the output value V6 is higher than the input value V7. Fig. 3B schematically shows a number of parameters S₅E₅D on which the mapping of Fig. 2B may be based. A third one 232 of the pixels of the input image has a value V5 which is lower than the minimum value 208 of the output range of values 202. The image conversion unit 100 according to the invention is arranged to compute the corresponding output value V8. Because the third one 232 of pixels of the input image corresponds to the lowest value of pixel values of the input image 110, the corresponding output value V8 is preferably equal to the minimum value 208 of the output range of values 202. Because, V5 is lower than the minimum value 208 of the output range of values 202 and because the spatial distance S between the third one 232 of the pixels of the input image and a fourth one 230 of the pixels of the input image is below a predetermined spatial threshold, the value V6 of the output pixel 236 corresponding to the fourth one 230 of the pixels of the input image is higher than the value V7 of the fourth one 230 of the pixels of the input image.

Preferably, the value V4 of the output pixel 236 corresponding to the fourth one 230 of the pixels of the input image is computed by multiplying the value V4 of the fourth one 230 of the pixels of the input image with the gain factor F , as specified in Equation 3.

Preferably, the gain factor F is also dependent on the actual exceed level E, i.e. the difference between the value V5 and the minimum value 208 of the output range of values 202. On basis of the actual spatial distance S and the actual exceed level E a gain factor F is determined by means of a iunction f₂(s,e) of spatial distance s and exceed level e . See Equation 6.

Preferably, the gain factor F is proportional to the exceed level e . See Equation 7.

Preferably, the gain factor F is also depends on the actual difference D between the value V5 and the value V7. On basis of the actual spatial distance S, the actual exceed level E and the actual difference D a gain factor F is determined by means of a function f₃(s,e,d) of spatial distance s , exceed level eand difference d . See Equation 8.

Preferably, the gain factor F is inversely proportional to the difference d . See Equation 9.

Fig. 4 schematically shows two exemplary functions f₃(s,e,d) for determining a gain factor F to be applied for the mapping of an input value to an output value. The x-axis corresponds with spatial distance s and the y-axis corresponds to the gain factor F . For a first one of the functions f₃(s,El,d) the actual exceed level equals El and for the second one of the functions f₃(s,E2,d) the actual exceed level equals E2. Fig. 5 schematically shows an embodiment of the image processing apparatus 500 according to the invention. The image processing apparatus 500 comprises: receiving means 502 for receiving a signal representing input images; an image conversion unit 504 for converting the input images into output images, as described in connection with any of the Figs. 1,2A,2B,3A,3B and 4; and an optional display device 506.

The signal may be a broadcast signal received via an antenna or cable but may also be a signal from a storage device like a VCR (Video Cassette Recorder) or Digital Versatile Disk (DVD). The signal is provided at the input connector 508. The image processing apparatus 500 might e.g. be a TV. Alternatively the image processing apparatus 500 does not comprise the optional display device 506 but provides the output images to an apparatus that does comprise a display device 506. Then the image processing apparatus 500 might be e.g. a set top box, a satellite-tuner, a VCR player, a DVD player or recorder. Optionally, the image processing apparatus 500 comprises storage means, like a hard disk or means for storage on removable media, e.g. optical disks. The image processing apparatus 500 might also be a system being applied by a film-studio or broadcaster.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be constructed as limiting the claim. The word 'comprising' does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements and by means of a suitable programmed computer. In the unit claims enumerating several means, several of these means can be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words are to be interpreted as names.

Claims

CLAIMS:

1. An image conversion unit (100) for converting an input image (110) into an output image (116), the output image (116) comprising output pixels having respective values in an output range of values (202), the image conversion unit comprising: a filter (102) for establishing that a first one (114,232) of the input pixels of the input image (110) has a first value which exceeds the output range of values (202); and a pixel processor (104) for assigning a second value to a second one (118,236) of the output pixels on basis of a third value of a second one (112,230) of the input pixels and on basis of a spatial distance between the first one (114,232) of the input pixels and the second one (112,230) of the input pixels, the second one (112,230) of the input pixels and the second one (118,236) of the output pixels having mutually equal spatial coordinates.

2. An image conversion unit as claimed in claim 1, wherein the pixel processor (104) is arranged to assign the second value to the second one (118) of the output pixels, the second value being lower than the third value of the second one of input pixels, if the first value of the first one (114) of the input pixels is higher than the maximum value of the output range of values (202).

3. An image conversion unit as claimed in claim 1 or 2, wherein the pixel processor (104) is arranged to assign the second value to the second one (236) of the output pixels, the second value being higher than the third value of the second one of input pixels, if the first value of the first one (232) of the input pixels is lower than the minimum value of the output range of values (202).

4. An image conversion unit as claimed in any of the claims above, wherein the filter (102) has an aperture (120) for establishing that the first one (114,232) of the input pixels of the input image (110) has a first value which exceeds the output range of values (202), which simultaneously covers the first one (114,232) of the input pixels and the second one (112,230) of the input pixels and which is substantially smaller than the size of the input image (110).

5. An image conversion unit as claimed in any of the claims above, wherein the pixel processor (104) is arranged to compute the second value of the second one (118,236) of the output pixels on basis of a first difference between the first value of the first one (114,232) of the input pixels and the maximum value of the output range of values (202) or the minimum value of the output range of values (202).

6. An image conversion unit as claimed in any of the claims above, wherein the pixel processor (104) is arranged to compute the second value of the second one (118,236) of the output pixels on basis of a second difference between the first value of the first one (114,232) of the input pixels and the third value of the second one (112,230) of the input pixels.

7. An image conversion unit as claimed in any of the claims 1-4, wherein the pixel processor (104) is arranged to compute the second value of the second one (118,236) of the output pixels by means of multiplying the third value of the second one (112,230) of the input pixels with a factor which is based on the spatial distance between the first one (114,232) of the input pixels and the second one (112,230) of the input pixels.

8. An image conversion unit as claimed in claim 7, wherein the factor is determined by a monotone function of spatial distance.

9. An image conversion unit as claimed in any of the claims 7-8, wherein the function corresponds to any function of the set of functions comprising: cosine, square cosine and Gaussian.

10. An image conversion unit as claimed in any of the claims 7-9, wherein the factor is based on a first difference between the first value of the first one (114,232) of the input pixels and the maximum value of the output range of values (202) or the minimum value of the output range of values (202).

11. An image conversion unit as claimed in any of the claims 7-10, wherein the factor is proportional to the first difference.

12. An image conversion unit as claimed in any of the claims 7-11, wherein the factor is based on a second difference between the first value of the first one (114,232) of the input pixels and the third value of the second one (112,230) of the input pixels.

13. An image conversion unit as claimed in any of the claims 7- 12, wherein the factor is inversely proportional to the second difference.

14. An image conversion unit as claimed in any of the claims 1-4, wherein the pixel processor (104) is arranged to compute the second value of the second one (118,236) of the output pixels by means of subtracting a term from the third value of the second one (112,230) of the input pixels, wherein the term is based on the spatial distance between the first one (114,232) of the input pixels and the second one (112,230) of the input pixels.

15. An image processing apparatus, comprising: - receiving means (502) for receiving an input image (110); and an image conversion unit for converting the input image (110) into an output image (116) (606), as claimed in any of the claims above.

16. An image processing apparatus as claimed in claim 15, further comprising a display device (506) for displaying the output image (116).

17. A method of converting an input image (110) into an output image ( 116), the output image (116) comprising output pixels having respective values in an output range of values (202), the method comprising: - establishing that a first one (114,232) of the input pixels of the input image

(110) has a first value which exceeds the output range of values (202); and assigning a second value to a second one (118,236) of the output pixels on basis of a third value of a second one (112,230) of the input pixels and on basis of a spatial distance between the first one (114,232) of the input pixels and the second one (112,230) of the input pixels, the second one (112,230) of the input pixels and the second one (118,236) of the output pixels having mutually equal spatial coordinates.

18. A computer program product to be loaded by a computer arrangement, comprising instructions to convert an input image (110) into an output image (116), the output image (116) comprising output pixels having respective values in an output range of values (202), the computer arrangement comprising processing means and a memory, the computer program product, after being loaded, providing said processing means with the capability to carry out: - establishing that a first one (114,232) of the input pixels of the input image