JPH0476550B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a circuit for extracting clear image signals from signals from a robot's eye, a surveillance camera for factory automation, or the like.

[Conventional technology]

Conventionally, as a method for obtaining clear images from blurred images, Proceedings of ICASSP 1981
The method described on pages 1117 to 1120 of 2010 is known. In this method, what is called a blurred image is the second
As shown in Figure a, in areas where the intensity is extremely weak or strong, the fine fluctuation signals of the image are not sufficiently recorded and are output as image signals due to the limitations of the dynamic range of the projector, so the preferred processing is As shown in FIG. 2b, the extremely weak or strong intensity portions are made larger or smaller, respectively, and the fine fluctuation signals are enhanced accordingly.

FIG. 3 is a block diagram of a conventional image enhancement circuit that implements the above method. The image signal input from terminal 1 is input to low-pass spatial filter 2, which reduces the average local intensity. A signal is output. On the other hand, the image signal inputted from the terminal 1 is inputted to the subtracter 3, where the average local intensity signal from the low-pass spatial filter 2 is removed, and only the detailed signal of the image is obtained. In order to emphasize the detail signals of this image, the average local intensity signal is transmitted to a multiplier 4 via a weighting circuit 5 which performs non-linear weighting.
In other words, the weighting circuit 5 is configured to output a large value in order to selectively amplify only the detail signal of the image corresponding to a portion where the average local intensity signal is extremely weak or strong. Furthermore, if the detailed signals of the image are emphasized and added to the average local intensity signal in this way, the dynamic range of the image signal may be exceeded, so the weighting circuit 6 corrects the average local intensity signal. Then, the corrected average local intensity signal and the enhanced image detail signal output from the multiplier 4 are added together by an adder 7 and outputted from an output terminal 8. As a result, parts with appropriate intensity can be output as they are, while parts with too strong or too weak intensity can be adjusted and output.
Image enhancement can be performed.

[Problem that the invention seeks to solve]

However, in the conventional method, the characteristics of the weighting circuits 5 and 6 differ from input image to input image because of the desire to leave parts with appropriate intensities as they are and the intensities have non-negative values. In addition, good results cannot be obtained unless non-linear characteristics are provided, and when processing still images such as photographs, the weighting circuits 5 and 6 must be carefully selected.
However, this is inconvenient for processing signals from television cameras in real time.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image enhancement circuit that can determine the characteristics of a weighting circuit regardless of the characteristics of an input image.

[Means for solving problems]

The image enhancement circuit of the present invention includes: a subtraction circuit that subtracts a predetermined offset value from an input image signal; a low spatial filter that performs low spatial frequency filtering on the output of the subtraction circuit; and an output of the subtraction circuit. means for removing the output of the low-pass spatial filter to obtain a high-pass signal; means for amplifying the high-pass signal according to the output of the low-pass spatial filter; and attenuating the output of the low-pass spatial filter. The present invention includes an attenuation circuit, and an adder that adds the output of the attenuation circuit, the amplified high-frequency signal, and the predetermined offset value.

[Effect]

The principle of the present invention is to allow the characteristics of the weighting circuit to be the same for many image signals by accepting some variation in the proper intensity. Specifically, by first subtracting an offset value that is approximately the center value of the dynamic range of the image signal, and then adding it after processing, the average intensity is made to be around the offset value. In this way, the output of the low-pass spatial filter is also subtracted by the offset value, so that both strong and weak intensity areas are areas where the distance between the offset and the average local intensity is large. In other words, the weighting circuit for amplifying the detail signal of the image may be proportional to the distance between the average local intensity and the offset value, and the weighting circuit may be one that is proportional to the distance between the average local intensity and the offset value. The weighting circuit of
It is sufficient to provide attenuation proportional to the distance between the average local intensity and the offset value. In reality, there is a problem with human vision, and the proportional coefficient differs slightly depending on the positive or negative side of the average local intensity signal from the offset value, and the proportional coefficient itself changes slightly depending on the distance from the offset value. Better characteristics can be obtained by changing the characteristics, but in any case, by doing this, the characteristics of the weighting circuit that depended on the input image were required in the past, but according to the present invention, the characteristics of the weighting circuit that does not depend on the input image can be achieved. Processing can be done using circuits.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

Figure 1 shows an example in which the image input signal is assumed to be displayed in 8 bits (256 gradations) and the offset value is set to 128, which is the median value.
Input terminal 1, subtracter 10, low-pass spatial filter 2,
Subtractor 3, weighting circuit 50, multiplier 4, adders 7, 90, offset input terminal 20, attenuation circuit 6
0 and an output terminal 8. As the low-pass spatial filter 2, the one shown in FIG. 1 described in 1985 IEICE Information Systems Division National Conference Lecture No. 168 can be used, and it includes a weighting circuit 50 and an attenuation circuit 60.
This will be discussed later.

The image signal applied from terminal 1 is subtracted by 128 as an offset by subtracter 10, and the result is passed through low-pass spatial filter 2 to obtain an average local intensity signal from which the offset has been subtracted. By subtracting this signal from the signal before passing through the low-pass spatial filter 2 with a subtracter 3, a detail signal of the image is obtained. On the other hand, the amplitude of the average local intensity signal from which the offset has been subtracted indicates a distance with a polarity sign from the offset, and this distance is converted into a value proportional to the positive distance by the weighting circuit 50 and then applied to the subtracter. A multiplier 4 amplifies and emphasizes the image detail signal output from the multiplier 3.
On the other hand, the average local intensity signal from which the offset has been subtracted is attenuated by an attenuation circuit 60, and this result and the enhanced image detail signal output from the multiplier 4 are added together by an adder 7. ,
An adder 90 adds the offset subtracted on the input side.

Since the weighting circuit 50 outputs a value depending on the amplitude of the polarity code signal, the circuit shown in FIG. 4 can be used. This weighting circuit has input terminal 1
00, an absolute value circuit 101, a multiplier 102, an output terminal 103, and a coefficient input terminal 104. A polarity sign signal inputted from the terminal 100 becomes a positive distance signal by the absolute value circuit 101, and the Since the signal is multiplied by the value added from and output from the terminal 103, the larger the absolute value distance signal, the larger the value obtained.

If you want the weighting circuit 50 to have the characteristic of changing the proportional coefficient depending on positive and negative input signals, or outputting a special value when the input signal is close to zero,
A ROM can be used instead. In this case, input the input signal to the address port and write the value you want to output to the address location corresponding to the input signal.

The configuration of the attenuation circuit 60 is as shown in FIG. 5, with an input terminal 200, a multiplier 201, and an output terminal 2
02 and a coefficient input terminal 203.
The average local intensity signal subtracted by the offset is input from the input terminal 200,
The multiplier 201 may attenuate the signal by the coefficient input from 3 and output it. Considering the later added offset, this attenuation has the effect of concentrating the average local intensity signal around the offset.

If nonlinear characteristics are desired for the attenuation circuit 60 as well, a ROM may be used as a substitute, similar to the weighting circuit 50.

〔Effect of the invention〕

As explained above, according to the present invention, it is possible to concentrate the average intensity around the offset value without being influenced by the characteristics of the input image, and at the same time to emphasize the detailed signals of the image.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing the principle of a conventional system, FIG. 3 is a diagram showing an example of a conventional image enhancement circuit, and FIG. FIG. 5 is a diagram showing an example of the weighting circuit, and FIG. 5 is a diagram showing an example of the attenuation circuit of FIG. 1...Input terminal, 2...Low-pass spatial filter, 3
...Subtractor, 4...Multiplier, 7...Adder, 8...
...Output terminal, 10...Subtractor, 20...Offset input terminal, 50...Weighting circuit, 60...Attenuation circuit, 90...Adder.

Claims (1)

[Scope of Claims] 1. A subtraction circuit that subtracts a predetermined offset value from an input image signal, a low-pass spatial filter that performs low-pass spatial frequency filtering on the output of the subtraction circuit, and an output of the subtraction circuit. means for removing the output of the low-pass spatial filter to obtain a high-frequency signal; means for amplifying the high-frequency signal according to the output of the low-pass spatial filter; and attenuation for attenuating the output of the low-pass spatial filter. An image enhancement circuit comprising: a circuit; and an adder that adds the output of the attenuation circuit, the amplified high-frequency signal, and the predetermined offset value.