JPS6348054A - Picture information inputting device - Google Patents

Abstract

PURPOSE:To smooth the steep difference of a desity and to prevent the deterioration of the quality of picture by outputting an output from either one of the overlapping two picture elements which is adopted according to a prescribed probability distribution regarding the overlapping part of plural line image sensors. CONSTITUTION:The picture element output is determined so that the adopting probability of the picture element due to the line image sensor 4 is high at the jointing part (c) of the line sensor 4, and said adopting probability comes gradually lower with an advance along a main scanning direction, and the adopting probability of the picture element of the line sensor 4 comes to be minimum at the jointing part (d), and besides, the adopting probability of the picture element of the line sensor 5 is low at the jointing part (c), and said adopting probability comes gradually higher with the advance along the main scanning direction, and the adopting probability comes to be maximum at the jointing part (d). When the picture element output signal of the overlapping part (c)-(d) determined in such the way is reproduced as the picture, the picture is smoothed visually and accordingly the quality of the picture is uniformized.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention relates to an image information human-powered device, particularly an image information human-powered device that manually generates image information for one line by arranging a plurality of line image sensors in a staggered manner. It is related to the device.

(Prior Art) Conventionally, in this type of image information human-powered device, the line image sensor for performing image information human-powered operation was switched by aligning the ends of the joints of each line image sensor as shown in FIG.

(Problems to be Solved by the Invention) However, there are variations in the output characteristics of each line image sensor.

Therefore, in the above configuration, band-like unevenness occurs in the sub-scanning direction on the reproduced image. In particular, when the original is an image with a density as shown in FIG. 3, it is assumed that the joining portion or switching boundary of the line image sensor is at position a. Then, the digital signal after the A/D conversion becomes as shown in FIG. 4 due to variations in the characteristics of each line image sensor. When this is reproduced as an image, a visual density difference as shown in FIG. 5 appears at the joint. When a density difference appears on the left and right sides of the joint in this way, this deteriorates the quality of the reproduced image, making it difficult to view the reproduced image.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned drawbacks of the conventional method and to perform image information processing with uniform image quality by eliminating density differences at junctions due to variations in output characteristics of each line image sensor.

[Means for solving problems]

The present invention includes a plurality of line image sensors arranged in the main scanning direction so that a plurality of pixels at the ends overlap with each other;
Output signals from each pixel other than the overlapping portion of multiple line image sensors are output as they are, and for the overlapping portion, an output signal from one of the two overlapping pixels adopted according to a predetermined probability distribution is output. and an output means for.

[For production]

According to the present invention, one line of pixel data from a plurality of line image sensors is stored in the storage means, and the pixel data of the overlapping portion of each sensor is read out according to a preset probability distribution, thereby eliminating density differences. Smoothing to prevent deterioration of image quality.

(Example) In the present invention, the line ease sensor is arranged as shown in FIG. In particular, the joint portion of the line image sensor 4.5 is schematically shown in FIG. The thick line represents the main body of the line image sensor, and the portion delimited by the thin line represents the image sensor for one pixel. In this case, the output digital signal after Δ/D conversion is as shown in FIG. The output of image sensor 4 is h
The output of the image sensor 5 is indicated by i. The starting end position of the line image sensor 5 in FIG. 6 is assumed to be 65, and the ending end position of the sensor 4 is assumed to be d. , C%d correspond to e and f in FIG. 7, respectively. Here, for example, when dividing by g, and using the digital signal for the section e-g and the digital signal i for the section g to f as the output corresponding to each pixel in the main scanning direction, it is the same as in Figure 4. (there is a step).

Therefore, in the interval from e to 8, more digital signals than the digital signal i are used as the output of the corresponding pixel.
Adopt more signals than signals. In the vicinity of Umari e, the proportion of digital signals adopted is large, and the proportion gradually decreases until it reaches the same proportion as that of digital signals i in the vicinity of g,
Conversely, as f approaches, the proportion of digital signal i gradually increases. This overlapped portion c− determined in this way
The pixel output signal of d is as shown in FIG. When this is reproduced as an image, the output pixels on the reproduced image are visually sufficiently small, so the image is visually smoothed and a visual density as shown in FIG. 9 is obtained. In other words, a visual density step as shown in FIG. 5b does not occur. In the present invention, as shown in FIG. 6, the line image sensor 4 is
Conversely, the probability of adopting the pixel output of the line image sensor 4 is high, and as it advances in the main scanning direction, the probability of adopting it gradually decreases. The probability of adopting the pixel output of the line image sensor 5 is low in the part C of , and the probability of adopting it gradually increases as it advances in the main scanning direction, and the probability of adopting the pixel output of the line image sensor 5 is the highest in the part d. Determine the output.

Next, a specific configuration of the present invention will be explained according to the drawings.

FIG. 1 is a block diagram of an image processing apparatus to which the present invention is applied. Reference numeral 17 denotes an image information human power device of the present invention, and in this embodiment, a case will be described in which the line image sensor section has three line image sensors. FIG. 10 shows details of the image information human power device 17 of the present invention.

As shown in Fig. 1θ, line image sensors 21 to 23
As described above, the pixels are arranged in a staggered manner so that the pixels overlap at the junction. The reflected light from the original is photoelectrically converted by line image sensors 21 to 23 and output as an analog density signal.

The A/D converters 24 to 26 each convert the analog density signal output from the corresponding line image sensor into a digital image signal and output the digital image signal.

The output digital image signals are sent to two line buffer memories 27 to 27 corresponding to each A/D converter.
32.

Two line buffer memories are provided for one line image sensor. Line buffer memories 27.29.31 and 28.30.32 are divided into two memory groups (here, the former is group A and the latter is group B). Input/output to the line buffer memory is performed alternately for each group. In other words, when writing data from the A/D converter to the line buffer memory of group A, reading from the line buffer memory of group B; Conversely, group B performs writing.

When outputting based on probability distribution, the output data of pixels in areas that do not overlap with each other is read out only from the data of pixels that fall on the corresponding sensor, and the output data of pixels in areas that overlap are read out to each sensor. Specify the address of the line buffer memory storing the data to be adopted (extract data) from the corresponding line buffer memory and read it. These processes are performed by the memory output control circuit 14.

The memory output control circuit 14 includes an address counter 33, an address memory 34, an address counter 35, and a selection circuit 36, as shown in FIG. Address memory 34 stores addresses for line buffers 27 to 32, and outputs the address of the line buffer memory set in advance by address counter 33. However, the read address that adopts the data stored in the address memory 34 and output to the output terminal 37 follows the above-mentioned adoption probability distribution. The address counter 35 outputs an address for writing the digital image signal data from each pixel of the line image sensor in the scanning order in the main scanning direction to the line buffer memory.

These memories 34 and counters 3 are selected by the selection circuit 36.
The address output from 5 is selected and output to each line buffer memory. The line buffer memories 27 to 32 are divided into the aforementioned groups A and B, and perform human output alternately.

Therefore, while the A group is writing data from the line image sensors 21 to 23 according to the output address data of the address counter 35, the B group is reading data according to the output address data of the address memory 34.

Note that at this time, data is written to each line buffer memory of the A group from each sensor with the same address designation. During this time, pixel data corresponding to the three line image sensors are read out from the line buffer memory of the corresponding B group in order in the main scanning direction. In other words, writing is performed in parallel for three line image sensors,
In contrast, reading is performed sequentially. Therefore, the counting clocks of address counters 33 and 35 are different.

The selection circuit 36 distributes and sends an address signal, a read/write switching signal, and a read permission signal to each line buffer unmemory. The line buffer memory receives and receives these signals and performs writing and reading.

In other words, the line buffer memory performs tFf5 writing in response to a write signal and address designation, and reads out in response to a read signal and address designation. Here, unless the line buffer memory receives the read permission signal, the stored data will not be read out, and the line buffer memory will be in a state of being disconnected from the circuit.

In this way, writing to and reading from the line buffer memory is performed for each group, and especially when reading out the pixel overlap area, signals are read from the line buffer memory only from read addresses set in advance as a probability distribution. It is read out and sent from the terminal 37 to the next image processing section 15.

In the embodiments described above, a case was described in which three line image sensors were used. When the number of line image sensors is further increased, it is sufficient to increase the number of line image sensors having joints at both ends like the line image sensor 22 shown in FIG. At this time, the same addressing may be repeated except for the line image sensors at both ends. Therefore, addressing is performed by repeatedly using the portion where addresses for the same addressing are stored. Therefore, the size of the address memory does not become very large.

Even if the number of line image sensors is increased in this way, it is only necessary to increase the number of A/D converters and line buffer memories corresponding to the number of line image sensors.

〔Effect of the invention〕

As explained above, according to the present invention, pixel data for one line from a plurality of line image sensors is stored in the storage means, and pixel data of the overlapping portion of each sensor is read out according to a preset probability distribution. This smoothes density differences and prevents deterioration of image quality. It is possible to smooth density differences at the junctions of a plurality of image sensors and prevent deterioration in image quality in human-powered image information.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of an embodiment of the present invention. FIG. 2 is a layout diagram of a line sensor arranged in a staggered manner with overlapping parts. FIG. 3 is an example of a density signal of an original image. The figure shows an example of the output for the human power in Figure 3, Figure 6 is an enlarged view of the joint in Figure 2, and Figures 7 to 9 are examples of the output of the embodiment. FIG. 10 is an example of an image processing device to which the present invention is applied;
The figure is a layout diagram of a conventional staggered line image sensor. 1 to 6... line image sensor, 11... line image sensor section, 12... A/D converter and decoder, 13... line buffer memory circuit, 14... memory input/output control circuit, 21-23... line image sensor, 24-26.
...A/D converter with decoder function, 27-32...
Line buffer memory, 33, 35...address counter, 34...address memory, 36...selection circuit, 37...output terminal. ■ し− −−+−−−−−−−−m−+ JFig. 3 Fig. 4 Fig. 5- Main change I same 1 Fig. 6 Fig. 7 Fig. 9

Claims (1)

[Claims] (1) Multiple line image sensors arranged in the main scanning direction so that multiple pixels at the ends overlap with each other, and output signals from each pixel other than the overlapping portion of the multiple line image sensors remain unchanged. and output means for outputting an output signal from either one of the two overlapping pixels adopted according to a predetermined probability distribution for the overlapping portion.