JPH06233112A - Black frame detecting method for microfilm reader - Google Patents

Abstract

PURPOSE:To accurately obtain a black frame by eliminating the influence of stains or the like by correcting the white/black of a central picture element on a scanning line with the matrix of peripheral picture elements. CONSTITUTION:The scanning lines for black frame detection at fixed intervals are inputted to a first binarizing means 56, and the other scanning lines are inputted to a threshold value detecting means 58. The means 58 is provided with a histogram preparing means 62 and a threshold value calculating means 64. The means 62 prepares a histogram by using the output signal of a line sensor 52 which reads the line excepting for the scanning line for black frame detection. The means 64 calculates a threshold value by using this histogram. The threshold value is used for detecting the black frame of the scanning line for black frame detection to be read later. The reading signal of the scanning line is binarized by the means 56 while using the threshold value provided by the means 58, and inputted through a white/black correcting means 65 to a black frame detecting means 66, and the black frame is detected. The white/ black of the picture element on the scanning line for black frame detection is corrected by using peripheral picture elements. Thus, the influence of dirt or flaw on the image near an original or on the film is eliminated and the black frame can be exactly obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a black frame detection method for a microfilm reader which detects a black frame formed around a document image on a negative microfilm.

[0002]

2. Description of the Related Art On a negative microfilm, a transparent non-image portion is formed so as to surround the outside of an image on a document. When printing out an image of an original from this microfilm, it is necessary to output a range larger than the original so that the entire range of the original does not deviate from the output range.

FIG. 6 shows the relationship between microfilm and print output. On the microfilm 10, images 12 and 14 of an original having a certain size are recorded as negative images. Since the range 16 to be printed out is fixed, if this range 16 is printed out, the print images 12A of the images 12 and 14 on the print papers 18 and 20,
Black frames 22 and 24 are generated outside 14A.

A method of automatically discriminating and masking the black frames 22 and 24 has been proposed. One of the methods is a black frame if the image signal continuously outputs a fixed number N (for example, 224) of black pixels when the image on the scanning line 26 at regular intervals (for example, 1 mm) is read by the line sensor. It is determined to be 22, 24. The fixed number N = 224 is a length corresponding to 14 mm on the original, for example.

In this method, as shown in FIG. 6A, a coordinate x in which N black pixels are consecutive is detected from a coordinate X in which a pixel on the scanning line 26 changes from a white pixel to a black pixel. The coordinate X = x−N, which is N pixels back from the coordinate x, is determined as a black frame. Similarly, the coordinates X _n , X _{n + 1} , X _{n + 2} , ... Of the black frame are sequentially obtained for the _nth , _{n + 1} , _{n + 2} , ... A black frame range that connects X _n , X _{n + 1} , ..., That is, a masking range is obtained.

[0006]

2. Description of the Related Art Here, depending on the manuscript, the image 12,
14 may be close to the black frames 22 and 24. In this case, in the method of discriminating the black frames 22 and 24 from the array of pixels on the black frame detection scanning line 26 as described above, there is a problem that the black frames 22 and 24 cannot be discriminated correctly by the images close to the black frames 22 and 24. there were. This problem also occurred when the film 10 near the edges of the images 12 and 14 had stains or scratches.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and it is a microscopic device capable of accurately discriminating a black frame when there is an image near the periphery of an original or when the film has dirt or scratches. An object of the present invention is to provide a black frame detection method for a film reader.

[0008]

According to the present invention, the object is to detect a black frame range from an array of pixels indicated by an image signal obtained by reading an image in the main scanning direction and binarizing the image with a predetermined threshold value. A matrix of a predetermined number of peripheral pixels with one pixel on the scanning line for detection as a central pixel is formed, and the central pixel is inverted in black and white when a fixed number exceeding half of the peripheral pixels is opposite to the black and white in the central pixel. The black frame detection method of the microfilm reader is characterized in that the black frame is then detected.

Here, it is desirable that the peripheral pixel matrix is formed by using the pixels of another adjacent black frame detection scanning line. Further, the peripheral matrix may be formed by using the pixels of the scanning line for non-black frame detection which is apart from the scanning line for black frame detection by a constant distance.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is an overall configuration diagram of an embodiment of the present invention, and FIG.
Is an operation flow chart showing the operation, FIG. 3 is a view showing a histogram, and FIGS. 4 and 5 are views showing an arrangement example of a peripheral matrix.

In FIG. 1, reference numeral 30 is a microfilm, which is wound from one of the two reels 32 and 34 to the other and runs. The light from the light source 36 is guided to the film 30 from below, and the transmitted light from the film 30 is projected onto the projection lens 38, the image rotation prism 40, and the reflection mirrors 42 and 44.
It is led to the screen 48 from the backside via 46. A vertically movable plate 50 is provided on the back surface of the screen 48 so as to be movable in the horizontal direction, and a CCD line sensor 52 is attached to the movable plate 50. Therefore, this line sensor 52
It is possible to read the projected image on the screen 48 by moving the movable plate 50 in the horizontal direction while reading the image incident on.

At the time of the first image reading, the output signal of the line sensor 52 is transmitted through the changeover switch 54 shown in FIG. 1 to the first binarizing means 56 and the threshold detecting means 5.
8 is input. When the image is read for the second time, it is input to the second binarizing means 60.

As described above, the scanning lines 26 (FIG. 4) for detecting the black frame are read at regular intervals (for example, 1 mm intervals). Therefore, other scanning lines 26 corresponding to the image reading density are provided between the scanning lines 26. There will be scan lines left. For example, when reading a 1 mm width with 16 scanning lines 26, there are 15 scanning lines between the black frame detection scanning lines 26. Black frame detection scanning line 26
Scan lines other than the above, or black frame detection scan lines 26
All scan lines including the threshold value D as described below.
Used to calculate _TH .

In this embodiment, the scanning lines 26 for detecting the black frame at regular intervals are input to the first binarizing means 56, and the other scanning lines are input to the threshold detecting means 58.

The black frame detecting means 58 has a histogram creating means 62 and a threshold value calculating means 64. The histogram creating means 62 creates the density histogram shown in FIG. 3 by using the output signal of the line sensor 52 that has read the scanning lines other than the black frame detection scanning line 26. This histogram shows the distribution of the number of pixels with respect to the density D of each pixel.

Here, the histogram creating means 62 uses the nth black frame detecting scanning line 26 (n) and the preceding (n)
A histogram is obtained by 15 scanning lines between the -1) th black frame detection scanning line 26 (n-1). The threshold value calculation means 62 uses this histogram to calculate the threshold value D _TH (n) used for the n-th black frame detection scanning line 26 (n).

For example, in the histogram shown in FIG.
The mountain P ₁ on the white side is formed by the transparent portion of the film 10 outside the images 12 and 14 in FIG. 6. Further, the mountain P ₂ on the black side is formed by the backgrounds in the images 12 and 14 and the black frames 22 and 24 around them. Therefore, the required image will be distributed between the two peaks P ₁ and P ₂ . The threshold value calculation means 64 finds the density D _TH at a fixed ratio (for example, 70%) with respect to the density D ₂ of the mountain P ₂ as a threshold value for binarization, and stores the result.

In the threshold value detecting means 58, the (n-1) th and nth black frame detecting scanning lines 26 (n-1), 2
The threshold value D _TH (n) is obtained as described above with 15 scanning lines between 6 (n). This threshold D
_TH (n) is used for black frame detection of the nth black frame detection scanning line 26 (n) to be read thereafter. At this time (n
A histogram may be created including the -1) th black frame detection scanning line 26 (n-1).

The first binarizing means 56 uses the threshold value D _TH (n) thus obtained by the threshold value detecting means 58 to read out the output signal of the n-th scanning line 26 (n). Is binarized. The binarized signal is converted into a black and white correction means 65.
Then, it is inputted to the black frame detecting means 66, and the black frame is detected and stored therein.

The black-and-white correction means 65 is for black-and-white correction of the pixels on the n-th black frame detection scanning line 26 (n) by using the peripheral pixels. For example, as shown in FIG. 4, the peripheral pixel matrix is formed by using the black frame detection scanning lines 26 (n-1) and 26 (n + 1) before and after the scanning line 26 (n). Instead of the black frame detection scanning lines 26 (n-1) and 26 (n + 1), non-black frame detection scanning lines 26a and 26b may be used which are adjacent to each other or separated by a constant distance as shown in FIG.

In FIGS. 4 and 5, scan line 26 is shown.
(N) A 3 × 3 matrix centered on one pixel above, that is, the central pixel G ₀ is a peripheral pixel matrix. White correction means 65, among the other pixels G ₁ ~G ₈ surrounding the center pixel G _0, to count the number of the same pixel as the black and white of the central pixel G _0.

When the central pixel G ₀ is a black pixel, a fixed number exceeding half of the peripheral pixels G _{1 to} G ₈ (for example, 5).
If each pixel is a black pixel, the central pixel G ₀ is left as a black pixel. On the contrary, if the constant number (five) or more of the peripheral pixels G _{1 to} G ₈ is a white pixel, the central pixel G ₀ is corrected to a white pixel. When the central pixel G ₀ is a white pixel, the peripheral pixels G _{1 to} G ₈ are similarly generated.
When more than a certain number of black pixels are black pixels, the central pixel G ₀ is corrected to a black pixel.

Thus, by correcting the black and white of the central pixel G ₀ by the number of black and white pixels of the peripheral pixels G _{1 to} G ₈ ,
A black frame can be made clear by blurring an image due to a character portion, dirt, or scratches included in the image.

The black frame detecting means 66 detects the black frames 22 and 24 (see FIG. 6) by various methods using the array of pixels on the black frame detecting scanning line 26 (n) thus corrected in black and white. For example, as described above, the scanning line 26 may be included in the range of the image 12 and the black frame may be detected because N black pixels continue in succession ((A) in FIG. 6).
reference). Further, as shown in FIG.
Since a% (for example, 30%) of (for example, 160) pixels are black pixels, the presence of a black frame is detected, and a coordinate X = x, which is Nb pixels back from the coordinate x at which the number of black pixels becomes a%.
-Nb may be determined as a black frame.

In this case, for the Nb pixel, the range of the N pixel is (1
-B): indicates the number of pixels up to the coordinate divided into the ratio of b,
It is desirable to set b≈a. Note that the above description has been given of the case where the scanning enters from the outside of the images 12 and 14 into the images 12 and 14, but when the scanning goes from the inside of the images 12 and 14 to the outside, the black pixel of the above description is displayed. The white pixels may be reversed. When the black frame is detected by various methods in this manner, the black frame range is determined by connecting the points of the black frame obtained for each scanning line 26, and the black frame range is stored.

As described above, by the first image reading,
The detection of the black frame and the detection of the threshold value D _TH (n) are performed. Then, at the time of the second image reading, the obtained threshold value D _TH (n) is used to perform binarization in the second binarizing means 60. That is, an image near the nth scan line 26 (n) is binarized by using the threshold value D _TH (n), and the threshold value D _TH in the vicinity is sequentially read from the memory according to the image reading area. Read and use.

The binarized output signal is the image processing means 6
8 is input. This image processing means 68 enlarges an image
Various spatial filtering processes such as compression, edge enhancement by differential processing, thinning processing, dither processing, etc. are performed.

The image signal subjected to such image processing is masked by the masking means 70 in an unnecessary area such as the black frames 22 and 24 (see FIG. 6) and is output to the output means 72. The output unit 72 is formed by, for example, a printer, and erases the masking range from the image-processed image signal to print out.

Next, the operation will be described with reference to FIG. When performing the first image reading (scan) (step 10
0), first, the order m of scanning lines is initialized to 1 (step 102). Then, it is determined whether or not this order m matches the black frame detection scanning line 26 (step 10).
4). That is, the black frame detection scanning lines 26 appear at regular intervals, for example, every 16 scanning lines. Therefore, when m is an integer multiple of 16I (where I is an integer), the black frame detection scanning line 26 is m16I. The scan lines of are for histogram detection.

Therefore, in the first m = 1 to 15, m ≠ 1
Since it is 6I, m = 16 (I-1) +1 to m = 16I-
A histogram is obtained from the scan lines up to 1 (steps 106, 108, 110, 112), and the threshold value D _TH (1) is calculated and stored (step 11).
4). When m = 16, the scanning line 26 (1) for black frame detection is the first, so the threshold value D _TH obtained immediately before this is
The first binarization means 56 is binarized using (1) (step 116). Black-and-white correction is performed by the black-and-white correction means 65 using this binarized signal.

That is, first, the black frame detection scanning line 26
(N) A matrix including a predetermined number (N × N−1) of peripheral pixels G _{1 to} G _NxN−1 centered on one pixel (center pixel G ₀ ) on the top is formed (step 118). Then, of the peripheral pixels G _{1 to} G _NxN-1 , the number of pixels a in which black and white are opposite to that of the central pixel G ₀ is obtained. This pixel number a is the number of peripheral pixels (N × N−
If it is larger than half of 1) (step 120), that is, if a> (N × N−1) / 2, the black and white of the central pixel G ₀ is inverted (step 122). a <(N × N-
If 1) / 2, do not reverse.

After correcting the black and white of the central pixel G _{0 in} this way, a black frame is detected and stored (step 12).
4). When the above operation is repeated and the processing is completed for the entire area of the image (steps 110 and 112), the second scan is started (step 126).

In the second scan, the threshold value D _TH obtained from the histogram in step 114 is used to scan the image.
It is binarized (step 128), and image processing is performed using this binarized image (step 130). Further step 1
The black frame range obtained in step 24 is masked and erased (step 132), and printed out (step 13).
4).

In the above embodiment, the peripheral pixel matrix for the pixel of the n-th black frame detection scanning line 26 (n) is the black frame detection scanning lines before and after it, that is, (n
-1) th and n + 1th black frame detection scanning lines 26
It was calculated from (n-1) and 26 (n + 1). However, according to the present invention, the (n-1) th and (n + 1) th scan lines 26
Instead of (n-1) and 26 (n + 1), the peripheral pixel matrix may be formed by using scanning lines which are apart from the scanning line 26 (n) not for black frame detection between them by a constant distance.

[0035]

As described above, according to the first aspect of the present invention, when scanning lines are read at regular intervals to detect a black frame,
The black and white of the central pixel on the scanning line for detecting the black frame is corrected by using the peripheral pixel matrix to eliminate the influence of stains and scratches on the image and the film near the periphery of the document. You can ask.

Here, as the peripheral pixel matrix, pixels on the black frame detection scanning line before and after the black frame detection scanning line including the central pixel can be used (claim 2). Further, the pixels of the scanning line for black frame detection which is apart from the scanning line for black frame detection including the central pixel by a predetermined distance may be used (claim 3).

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of an embodiment of the present invention.

[Fig. 2] Operation flow chart

FIG. 3 is a diagram showing a histogram.

FIG. 4 is a diagram showing a peripheral pixel matrix.

FIG. 5 is a diagram showing another example of a peripheral pixel matrix.

FIG. 6 is an explanatory diagram of the relationship between microfilm and print output.

[Explanation of symbols]

10, 30 Microfilm 12, 14 Image 18, 20 Print paper 26 Black frame detection scanning line 52 Line sensor as image reading sensor 65 Black and white correction means 66 Black frame detection means 68 Image processing means 70 Masking means G ₀ Central pixel G ₁ ~ G ₈ peripheral pixels

Claims (3)

[Claims] 1. A method for detecting a black frame range from an array of pixels indicated by an image signal obtained by reading an image in the main scanning direction and binarizing the image with a predetermined threshold value, and selecting one of the black line detection scanning lines. Forming a matrix of a predetermined number of peripheral pixels having a pixel as a central pixel, and inverting the central pixel in black and white when a constant number exceeding half of the peripheral pixels is opposite to the black and white, and then detecting a black frame; A method for detecting a black frame of a microfilm reader, characterized by: 2. The black frame detecting method for a microfilm reader according to claim 1, wherein the peripheral pixel matrix is formed by using pixels of the black frame detecting scanning line and other black frame detecting scanning lines adjacent thereto. 3. A black frame detection scanning line is set at regular intervals through a predetermined number of scanning lines, and pixels of one black frame detection scanning line and another scanning line which is separated from this by a constant interval are used. The method for detecting a black frame of a microfilm reader according to claim 1, wherein a peripheral pixel matrix is formed by using the above method.