JPH07105890B2 - Image signal processing method - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image signal processing method for binarizing an image signal obtained by scanning a film projection image with an image sensor and then binarizing the image signal. It is a thing.

(Technical background of the invention) A film projection image is read by an image sensor such as a CCD line sensor or a CCD area sensor, and after this image signal is defocused, it is a binarization level that floats according to a certain binarization level or image signal. A method of comparing with the level has already been proposed (see Japanese Patent Laid-Open No. 62-130068).

The blur correction here is to emphasize the edges of the image by emphasizing the high frequency components of the image, and a high-frequency emphasis filter using a first derivative or Laplacian (second derivative), such as an unsharp mask (Unsharp Mask, Hereinafter referred to as US mask) is used. However, since such a high-frequency emphasis filter essentially uses differentiation, it is vulnerable to noise, and in particular, noise in the background region of the image signal is also emphasized, which significantly reduces the image quality after binarization processing. There was a problem of letting it. This problem becomes particularly important when reading a projected image of a film which is easily damaged by dust, dust or scratches.

Therefore, it has been considered by the applicant of the present application that the image signal is sliced at a predetermined background noise cut level close to the background level to remove noise included in the background. However, in this case, since the background density level differs depending on the image, how to determine the background noise cut level becomes a problem. That is, if this setting is wrong, background noise cannot be sufficiently removed, or the fluctuation range (dynamic range) of the effective signal level of the image signal is narrowed, resulting in deterioration of image quality.

Therefore, it is conceivable to determine this background noise cut level by using a histogram. In this case, if the image of the platen is included in a region other than the original, such as the periphery of the original, in the film projection image, the image of the original The region will create a different mountain in the histogram than the background mountain.
For this reason, it is possible that the background mountain cannot be correctly identified and the background noise cut level is set to an incorrect level.

(Object of the Invention) The present invention has been made in view of the above circumstances, and while a projection image of a film is read to obtain an image signal, the image signal is sliced at a background noise cut level close to the background level. When noise in the ground is removed, blurring is corrected by high-frequency emphasis, and then binarization processing is performed, the background noise cut level can always be set appropriately without being affected by the image signal for areas other than the original. An object of the present invention is to provide an image signal processing method capable of improving the image quality after binarizing a projected image.

(Constitution of the Invention) According to the present invention, an object of the present invention is to reduce an image signal obtained by scanning a projected image of a film, which is obtained by placing an original document smaller than the original document on a document table, to reduce background noise close to the background level. After slicing according to the level to remove background noise, blur correction is performed to sharpen the image signal, and in the image signal processing method that binarizes with a predetermined binarization level, the portion in which the original is imprinted From the histogram obtained from the image signal obtained by scanning a part of the projected image of the film so as to include, obtained by scanning a part of the projected image of the film so as not to include the part in which the original is imprinted Subtract the histogram obtained from the image signal, detect the mountain with respect to the background using the histogram after this subtraction, This is achieved by an image signal processing method, characterized in that the background noise cut level is set near the skirt on the side of the signal level including the image.

(Principle) FIG. 8 is a view showing the principle of the present invention, and FIG. 8A is a projected image of a film. The document is placed on a document table wider than this and shot on film. In FIG. 8A, reference numeral 1 indicates a portion on which the original is imprinted, and portion on which the two manuscript table is imprinted. This film shall be a negative film. In the projected image of this film, one scanning line is read by the licensor so that the portion 1 in which the original is imprinted is not included. That is, the portion indicated by O in FIG. 8A is scanned.
Since the film is a negative film, the platen becomes dark and the amount of light incident on the line sensor is small, so the histogram p at this time is one peak q located on the low output level side as shown in FIG. 8B. Will have.

Next, the original is scanned so as to include the portion 1 in which the original is imprinted. For example, scanning is performed at a position indicated by r in FIG. 8A to read one scanning line. At this time, one scanning line includes a portion on which the document table is imprinted and a portion on which the document is imprinted. This histogram s includes a peak q'for the portion in which the platen is imprinted as shown in FIG. 8C and a peak t for the background of the original. The density level V of the mountain q'is the same as the density level V of the mountain q shown in FIG. 8B.

The present invention subtracts the histogram p of FIG. 8B from the histogram s of FIG. 8C to obtain (s−αp). Here, a is a constant. Then, the background noise level is obtained using the histogram after this subtraction.

(Embodiment) FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a conceptual diagram of a part of a concrete example thereof, and FIGS. 3A to 3C are diagrams showing signal waveforms during processing, and FIG. Shows a 3 × 3 matrix, and FIG. 5 shows a histogram for a negative film.

In FIG. 1, reference numeral 10 is a light source, and the light from the light source 10 is guided to an image sensor 20 via a condenser lens 12, a film 14, a projection lens 16, and a mirror 18, and a projected image of the film 14 is directed to the image sensor 20. Form an image. The image sensor 20 is formed by a CCD line sensor, a CCD area sensor, or the like, is driven by a pulse supplied from a pulse circuit (not shown), scans an image, and outputs a time-series image signal a.

This image signal a has an output waveform as shown in FIG. 3A when the film 14 is negative. 3A to 3C, the horizontal axis represents time or pixel order, and the vertical axis represents voltage.

An image signal a for one scanning line for an area (o in FIG. 8A) that does not include the portion in which the original is imprinted is temporarily stored in a semiconductor memory 22 such as a line memory, and a histogram p of this area is displayed.
(FIG. 8B) is obtained by the histogram means (I) 24 constituted by a CPU (not shown). Next, the area to be scanned is changed to the area r (FIG. 8A) including the portion in which the original is imprinted, and the histogram means (II) 26 of the CPU obtains the histogram s of this area. The CPU subtracts the histogram p from the histogram s by the built-in subtraction means 28. At this time, if necessary, the level of the histogram p is changed by the subtractor 30 to be ap. Where a is a constant
It should be determined by the size of peaks q and q'in FIGS. The CPU obtains the background noise cut level b from the histogram (s-αp) after the subtraction by the background noise cut level detection means 32 of the CPU itself.

In the histogram, the horizontal axis represents the output level V of the image signal a and the vertical axis represents the frequency H. The histogram (s-αp) after subtraction with respect to the negative film 14 shows the peak B corresponding to the background on the low output level side. Will have. The CPU determines the background noise cut level b near the skirt on the right side (high output level side) of the mountain B in this histogram. For example, assuming that the total number of pixels of the histogram b is N, the output level having a high frequency of N / 100 can be determined as the background noise cut level b. The background noise cut level b can be set by various setting methods using the slope of the histogram, the distance D (FIG. 5) from the intersection C at which the peak B of the histogram intersects the height of a certain frequency, and the like.

The image signal a and the background noise cut level b are input to the background noise cut circuit 34, and the noise included in the background area is removed here. That is, this circuit 34 comprises, for example, a comparator 36 and a switch 38, as shown in FIG. 2, and the comparator 36 outputs the image signal a and the background noise cut level b.
Compare with. In addition, the switch 38 has the comparator 36 a <b
When it is judged that the background noise cut level b is selected,
When ≧ b, the image signal a is selected. This results in this circuit
The output c of 34 is sliced at the background noise cut level b and the noise in the background region is cut, as shown in FIG. 3B. In this way, the background noise cut level b is set close to the background level d (FIG. 3A) and slightly higher than this.

A blur correction circuit 40 emphasizes the high-frequency component of the signal c by using the US mask 42 as a high-frequency emphasis filter to emphasize the edge of the image.

The US mask 42 functions to emphasize the center pixel when the image space is an odd matrix in which the center pixel appears. As the US mask 42, for example, as shown in FIG.
When i is set to i, the data e for the central pixel is
The pixel data is used to convert to E = 5e− (b + d + h + f), and this E is used as a new image signal. In this case, the US mask 42 is set as shown in FIG. 2, and each element of this matrix is integrated with the image data of four surrounding pixels centering on the central pixel, and the sum E of the integrated values is obtained. The image signal e thus emphasized is compared with the binarized level f in the comparator 44 to obtain the binarized signal g.

In this way, the histogram of the area not containing the original is subtracted from the histogram of the area containing the original, and the background noise cut level is determined from the histogram after this subtraction, so the background in the histogram is not affected by the original table or the like. The ground peak can be correctly identified, and then the background noise cut level b can be correctly determined from the background peak.

FIG. 6 is a block diagram of another embodiment, and FIGS. 7A to 7D are output waveform diagrams of respective parts.

In this embodiment, the binarization level is changed by using the image signal a, that is, the image sensor 20.
Image signal a is stored in memory and smoothed by a smoothing circuit 50 to obtain a smoothed signal h (FIG. 7A).
Is compressed by the compression circuit 52. This compressed signal i (Fig. 7B)
The voltage level of is further raised by k by the level shift circuit 54. This level-shifted signal j (Fig. 7C)
Is compared with the signal d which has already been subjected to the blur correction in the comparator 44 to obtain the binarized signal l.

As the smoothing circuit 50 of this embodiment, it is possible to use, for example, a median filter that adopts an intermediate value (median) of a 3 × 3 matrix centered on the center pixel as image data of the center pixel.

According to this embodiment, since the binarization level j itself changes depending on the image signal a, it is suitable for highly accurate binarization processing of an image having a small contrast.

In each of the above embodiments, the US mask 42 is used for the blur correction circuit, but the present invention is not limited to this.

Further, the background noise cut level b is set to be slightly higher than the background d when the image of the negative film is used, but when the positive film is used, it is slightly lower than the background d. Needless to say, it is set to a low value in this case, and in this case, it is set to the skirt on the left side (low output level side) of the histogram mountain.

(Effects of the Invention) As described above, the present invention shows a histogram obtained by scanning a projected image of a film obtained by photographing a document on which a document table is placed so as not to include a portion in which the document is imprinted. Subtract from the histogram obtained by scanning to include the embedded part, find the mountain for the background from the histogram after this subtraction, and set the background noise cut level near the skirt of this mountain. Even if the document table is included in the film projection image, the background peak can be correctly identified, and the background noise cut level can be always set to an appropriate level even if the film projection image changes. Therefore, in the binarization process after blur correction, the background noise cut level is too close to the background to remove noise sufficiently, or conversely, it is too far from the background level to sacrifice the dynamic range of the image signal. It is possible to perform high-quality binarization processing without causing inconveniences such as.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a conceptual diagram of a specific example thereof, FIGS. 3A to 3C are diagrams showing signal waveforms in a processing process, and FIG. 4 is a 3 × 3 matrix. FIG. 5, FIG. 5 is a diagram showing a histogram, FIG. 6 is a block diagram of another embodiment, FIGS. 7A to 7D are output waveform diagrams of respective portions, and FIG. 8 is a principle explanatory diagram of the present invention. 20 …… image sensor, 34 …… background noise cut circuit,
40 …… Bokeh correction circuit, 42 …… Unsharp mask, 44…
Comparator, a ... Image signal, b ... Background noise cut level, d ... Background level, f, j ... Binarization level.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G06F 15/64 400 L

Claims (6)

[Claims] 1. A background noise is obtained by slicing an image signal obtained by scanning a projected image of a film, which is obtained by placing a document smaller than this on a document table, at a background noise cut level close to the background level. After removing, the blur is corrected to sharpen the image signal,
Further, in the image signal processing method of binarizing by a predetermined binarization level, from the histogram obtained from the image signal obtained by scanning a part of the projected image of the film so as to include the part in which the original is imprinted, Subtract the histogram obtained from the image signal obtained by scanning a part of the projected image of the film so that it does not include the part in which the original is imprinted, and detect the peak with respect to the background using the histogram after this subtraction. An image signal processing method, wherein the background noise cut level is set in the vicinity of a skirt on a signal level side including an image among hills on the left and right of the mountain. 2. The image signal processing method according to claim 1, wherein the blur correction is performed using an unsharp mask. 3. The binarization level is obtained by smoothing the image signal, compressing the smoothed signal, and further performing level shift. Image signal processing method. 4. The image signal processing method according to claim 3, wherein the smoothing is performed using a median filter. 5. The projection image is a projection image obtained by a negative film, and the background noise cut level is set near the skirt of the high output level side of the mountain of the histogram with respect to the background. The image signal processing method according to claim 1. 6. The projection image is a projection image obtained by a positive film, and the background noise cut level is set in the vicinity of the skirt on the low output level side of the mountain of the histogram with respect to the background. The image signal processing method according to the first item.