JP2006279513A - Digital image compression and decompression device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To save labor costs by automatically inspecting picture quality obtained by a digital image compression circuit and shortening an inspection time. <P>SOLUTION: A digital image compression and decompression device is equipped with compression/decompression processing circuit 4 and 5 which compress and expand a digital moving picture signal based upon set parameters, a CPU 1a which detects a difference between a digital image automatically compressed/decompressed to obtain desired picture quality deterioration precision and the original image and setting the parameters so that the difference is less than a specified value, and a memory 7 in which the image is compressed based upon the parameters set by the CPU and recorded. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention includes a compression circuit for compressing a digital moving image signal based on set parameters and recording it on a recording medium, a decompression circuit for decompressing data compressed by the compression circuit, and a compression circuit and a decompression circuit. The present invention relates to a digital image compression / decompression apparatus that is used in an AV device, a multimedia device, or the like having a control unit for controlling and having a recording / playback function.

  An irreversible digital image compression / decompression device records images without significant degradation of image quality, such as a security camera, because the image quality after compression / decompression is significantly degraded in images that contain a lot of high-frequency components. In the case of application to an apparatus that needs to be performed, it is necessary to adjust the compression rate so that the image quality after compression / expansion does not deteriorate more than before compression during product inspection in the production line process. By the way, in Patent Document 1 below, when an image is compressed and recorded on a recording medium, the image after MPEG compression / expansion and the image before compression are displayed at the same time, thereby compressing the image while confirming the image quality. A personal computer that enables the above is disclosed. Japanese Patent Application Laid-Open No. 2004-228561 discloses an apparatus that controls the code amount of compressed image data.

  FIG. 5 shows a conventional digital image compression / decompression apparatus, which includes a CPU 1 as a control unit, a bus IF 2, an external recording media interface 3, compression / decompression processing circuits 4, 5, a selector 6, a memory 7, an NTSC / PAL encoder 8, and a bus. 9 The bus 9 is connected to a bus IF 2, an external recording medium interface 3, compression / decompression processing circuits 4, 5, and a selector 6. The external recording medium interface 3 is connected to an external recording medium such as a DVC, DVD, HDD, or magnetic recording device. 10 is connected.

  In FIG. 5, at the time of compression processing, the input video signal is an NTSC signal (number of effective pixels = 720 pixels × 480 lines, frame rate = 30 frames / sec, etc.), other variables such as a bit rate after compression and an intra interval. When set to 6 Mbps, 15 frames, etc., the video signal input from the video input terminal is stored in the memory 7 via the selector 6. The video signal stored in the memory 7 is read from the memory 7 at any time as instructed by the CPU 1 and is stored in a temporary storage memory (not shown) in the CPU 1 via the selector 6 and the bus IF 2. The data is sent to the compression processing circuit of any of the compression / decompression processing circuits 4 and 5 via the bus IF 2 and subjected to compression calculation processing by the compression processing circuit. The compressed data is recorded on the external recording medium 10 via the external recording medium interface 3.

In the decompression process, the compressed data read from the external recording medium 10 is stored in the memory 7 via the bus IF 2, the CPU 1 (temporary storage memory in the CPU 1), the bus IF 2, and the selector 6 in accordance with an instruction from the CPU 1. Next, the compressed data is sent to the decompression processing circuit of the compression / decompression processing circuits 4 and 5 via the memory 7, selector 6, bus IF2, CPU1 (temporary storage memory in the CPU1), and bus IF2. It is decompressed. The decompressed video signal is accumulated in the memory 7 via the bus IF2, CPU1 (temporary storage memory in the CPU1), bus IF2, and selector 6, and then externally as an analog video signal by the NTSC / PAL encoder 8. By outputting, it can be used for image confirmation on a monitor.
JP 2001-28749 A (abstract) Japanese Patent No. 2970663 (FIG. 1)

  However, the technology described in Patent Document 1 is applied in the final product shipment inspection in the production line process, and the input image and the image obtained by compressing / decompressing the input image are switched, and both images are monitored by humans, or In the case of performing an inspection method in which a final inspection is performed by measuring an error rate using a device and a product is shipped, a check is performed on an analog output image. Therefore, in the case of NG, it is difficult to judge whether an error is caused by analog or digital (before or after the encoder), and there is a problem that inspection time becomes longer and labor cost becomes higher. . In the technique described in Patent Document 2, it is not possible to determine how much the image quality is deteriorated because only the amount of code is controlled. In the case of an image containing a lot of high frequency components, a clear image may not be left.

  The present invention provides a digital image compression / decompression device capable of automatically inspecting the image quality by a digital image compression circuit in view of the problems of the above-described conventional example, and thus reducing the inspection time and labor costs. The purpose is to do.

To achieve the above object, the present invention compresses a digital moving image signal based on a set parameter and records it on a recording medium, and a decompression circuit that decompresses data compressed by the compression circuit. And a digital image compression / decompression apparatus comprising a control unit for controlling the compression circuit and the decompression circuit so as to maintain a predetermined image quality by the compression / decompression,
The control unit detects a difference between the digital moving image signal and the digital image signal compressed by the compression circuit and decompressed by the decompression circuit, and sets the parameter so that the difference becomes a predetermined value or less. Set the compression circuit,
The compression circuit compresses the digital moving image signal based on a parameter set by the control unit and records the compressed signal on the recording medium.

  According to the present invention, the difference between the digital moving image signal and the compressed / decompressed digital image signal is detected, and the parameters are set in the compression circuit so that the difference is not more than a predetermined value. Can be automatically inspected, and thus the inspection time can be shortened and the labor cost can be reduced.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a digital image compression / decompression apparatus according to the present invention, FIG. 2 is a block diagram showing a compression processing circuit of the compression / decompression processing circuit of FIG. 1, and FIG. FIG. 4 is a flowchart for explaining the processing of the CPU in FIG. 1.

  The circuit configuration shown in FIG. 1 is different from the conventional device shown in FIG. FIG. 2 shows an example of a compression circuit (encoder) with predictive coding in the compression / decompression processing circuits 4 and 5. The image signal to be compressed is applied to the frame / field DCT unit 13 via the bus IF 2, the field / frame rearrangement unit 11, and the subtractor 12, and the frame / field DCT unit 13 outputs 2 in units of 8 pixels × 8 pixels. Dimensional discrete cosine transform (DCT) is performed, and then the quantization unit 14 performs quantization (quantization is the number of small values by dividing the entire DCT transform coefficient by a certain value based on the control of the rate control unit 15. (The mechanism for reducing the code amount by expressing) is performed, and the compression is performed, and the compressed data is output to the bus IF2. In addition, inverse quantization is performed according to MPEG rules for intra (intra-frame coding) and non-intra (inter-frame coding), an inverse quantization unit 16, an inverse DCT unit 17, an adder 18, a prediction memory 19, and a frame / It is calculated by the field motion compensation prediction unit 20 (detailed explanation is omitted).

  The rate control is possible by adjusting the data rate control coefficient of the rate control unit 15, and the encoding control is performed by assigning each code amount of I, P, and B pictures and controlling the quantization parameter. For example, in the case of a system that controls the code amount in units of frames, the result of the code amount of the compressed data, the current and past intra-interval code amounts, and the code amount considered to be a free area (predicted to be usable) are compared. Thus, the rate control is performed by adjusting the data code amount of the frame currently being processed.

  FIG. 3 shows an example of the decompression processing circuit of the compression / decompression processing circuits 4 and 5, and includes an inverse quantization unit 21, an inverse DCT unit 22, an adder 23, a prediction memory 24, a motion compensation prediction unit 25, and the like. The process performs the reverse operation of the compression process to generate video output data (detailed explanation is omitted).

  Next, the operation of the CPU 1a will be described with reference to FIG. In step S1, the original image stored in the memory 7 is compressed / decompressed, and the image after the compression / decompression processing is stored in the memory 7. In step S2, a part of the original image stored in the memory 7 is sucked into the temporary storage memory in the CPU 1 via the selector 6 and the bus IF2, and in step S3, the data is stored in the memory 7. The image after compression / decompression processing at the same position as the original image is taken up into the temporary storage memory in the CPU 1 via the bus IF2. In subsequent steps S4 and S5, the original image at the same position and the image after compression / decompression processing are compared to detect a signal difference, and in step S6, a compression / decompression processing circuit is automatically set so as to achieve a desired image quality degradation accuracy. Adjust the rates of the compression circuits 4 and 5.

As a method of comparing the difference between the two in steps S4 and S5, for example, a root mean square difference can be obtained. Assuming that the number of pixels to be compared is n, the image data before conversion is Z0i, and the image data after compression / decompression processing is Zi, the difference between the pixels is ξ and the mean square difference is s ² .
ξ = Zi−Z0i
s ² = Σξ ² / n
And the smaller the value of s ^2, the smaller the difference. Therefore, a threshold TH of a desired difference level is set in advance, and the rate of the compression circuit of the compression / decompression processing circuits 4 and 5 can be automatically adjusted so that s ² ≦ TH. If s ² ≦ TH in step S5, the process may not be terminated as it is, but may be adjusted so that s ² approaches TH as much as possible. The decompressed video signal is stored in the memory 7 via the bus IF 2, CPU 1 (temporary storage memory in the CPU 1), bus IF 2, selector 6, and externally as video data via the NTSC / PAL encoder 8. By outputting, it is possible to simultaneously check the image on the monitor.

Here, the technique of Patent Document 2 is control to fit within a desired code amount, but the present invention aims to adjust the code amount so that it falls within a desired difference level. Therefore, it is possible to quantitatively determine how much the image quality after compression / decompression processing deteriorates by simultaneously controlling the code amount and the difference comparison, and even in the case of an image containing a lot of high frequency components, Since an image with high image quality can be recorded, it can be used for an apparatus that needs to record image data without reducing the image quality, such as a security camera. In addition, by providing this function to the CPU 1a, it becomes easier to specify the cause of image quality degradation.
(1) Reduction of final inspection time for product shipment (2) Labor cost for final inspection of product shipment is reduced, and cost reduction can be realized.

1 is a block diagram showing an embodiment of a digital image compression / decompression apparatus according to the present invention. FIG. 2 is a block diagram illustrating a compression processing circuit of the compression / decompression processing circuit of FIG. 1. FIG. 2 is a block diagram showing an expansion processing circuit of the compression / decompression processing circuit of FIG. 1. It is a flowchart for demonstrating the process of CPU of FIG. It is a block diagram which shows the conventional digital image compression / decompression apparatus.

Explanation of symbols

1a CPU
4, 5 Compression / decompression processing circuit 10 External recording media

Claims (1)

A compression circuit for compressing a digital moving image signal based on a set parameter and recording it on a recording medium, a decompression circuit for decompressing data compressed by the compression circuit, and controlling the compression circuit and the decompression circuit A digital image compression / decompression apparatus that maintains a predetermined image quality by the compression / decompression.
The control unit detects a difference between the digital moving image signal and the digital image signal compressed by the compression circuit and decompressed by the decompression circuit, and sets the parameter so that the difference becomes a predetermined value or less. Set the compression circuit,
The digital image compression / decompression apparatus, wherein the compression circuit compresses the digital moving image signal based on a parameter set by the control unit and records the compressed signal on the recording medium.

Images