KR20030028653A - Apparatus and method for capturing image signals based on significance and apparatus and method for compressing and de-compressing significance-based captured image signals - Google Patents

Abstract

PURPOSE: A system and a method for collecting video signals based on significance, a system and a method for compressing and decompressing video signals collected based on significance are provided to reduce data volume of the collected video signals. CONSTITUTION: A system for collecting video signals based on significance includes a video signal collector(210), a significance calculator(250), and a video processor(260). The video signal collector collects a video signal with a predetermined resolution at a predetermined frame rate. The significance calculator calculates significance for the collected video signal. The video processor receives the video signal, processes the video signal according to the significance and generates collection information according to the processed video signal. The video processor ignores the video signal and records it in the collection information when the significance is lower than the first threshold value. When the significance is lower than the second threshold value larger than the first threshold value, the video processor reduces the resolution of the video signal and records it in the collection information.

Description

Apparatus and method for capturing image signals based on significance and apparatus and method for compressing and de-compressing significance-based captured image signals}

TECHNICAL FIELD The present invention relates to image processing, and in particular, to a method and apparatus for collecting an image signal based on importance and a method and apparatus for compressing and decompressing a collected image signal.

As digital electronics has evolved, many changes have been made to life. In particular, the development of digital technology has led to the development of multimedia technology, and as the Internet network becomes faster, users can enjoy vast amounts of multimedia information in real time.

Multimedia information involves a huge amount of data. However, as the amount of data increases, hardware limitations increase in transmitting or storing multimedia information through a network. Accordingly, various techniques for compressing multimedia information such as voice information or video information have been developed.

A representative compression standard is the Moving Picture Experts Group (MPEG). MPEG represents several conventions for compressing video signals, compressing audio signals, and compressing signals that combine video and audio signals.

MPEG-1 is designed to meet the data rate of 1.5Mbps, which is the CD-ROM driver's data rate, and MPEG2 is designed to be suitable for broadcasting and high-speed network environments. Therefore, when the broadcast television data is compressed, the algorithm is determined so that the maximum bit rate is 4 (Mbps).

In compression techniques such as the MPEG moving picture compression algorithm, all frames are regarded as separate purified images, and the similarity is used between adjacent frames. Techniques that exploit these similarities include prediction and interpolation.

In MPEG video, frames are called I, P, and B frames, and an I-frame (Intra-coded frame) is a technique of removing spatial redundancy from information of the current frame without reference to adjacent frames. To eliminate spatial redundancy, methods such as DCT and wavelet may be used.

Predictive-coded frame is based on the fact that image blocks move sideways, and is used to remove temporal redundancy by referring to previous I frame information and information of P frame, and B frame is used before and after I. Since the temporal redundancy is eliminated by using both and P frames, a high compression ratio can be obtained.

As these compression techniques are developed, it is possible to transmit a large amount of images, and digital multimedia technologies are used in various fields.

Digital multimedia technology has the following advantages over analog methods. First, sequential access is possible for each frame. In addition, as in the analog system, image quality deterioration due to repetitive reproduction does not occur. In addition, it is possible to control remotely using digital technology and to perform image processing.

As a representative example of such a digital multimedia technology, there is a field of a digital video recorder (DVR). A DVR is a device that converts analog video signals into digital video signals and records them. By converting and compressing an image digitally, image quality can be maintained, image processing can be performed, and sequential access can be performed for each scene.

However, the problem of increasing compression efficiency in compressing digital data has not yet been solved. As the amount of video information to be stored increases, a method for efficiently compressing video information is urgently required. Furthermore, DVR technology is now widely used in the field of security equipment. However, the security equipment records video signals transmitted from multiple cameras, and since the long-term record keeping is essential, the amount of video signals to be stored is huge. In order to record a large amount of video signals, hardware expansion required is required, so increasing the compression ratio of the video signals may reduce hardware costs.

Therefore, there is an urgent need for a technique capable of increasing the compression ratio of a video signal.

It is a first object of the present invention to provide an image signal collection apparatus capable of reducing the amount of data of an image signal to be collected.

It is a second object of the present invention to provide a video signal collection method capable of reducing the data amount of a video signal to be collected.

It is a third object of the present invention to provide an apparatus for compressing an image signal having a reduced amount of data.

It is a fourth object of the present invention to provide a method for compressing an image signal having a reduced amount of data.

It is a fifth object of the present invention to provide an apparatus for decompressing a compressed video signal having a reduced amount of data.

A sixth object of the present invention is to provide a method for decompressing a compressed video signal having a reduced amount of data.

1 is a view showing a video signal collecting device according to the prior art.

2 is a block diagram schematically illustrating an apparatus for collecting video signals according to an embodiment of the present invention.

3 is a flowchart illustrating a video signal collection method according to an embodiment of the present invention.

4 is a view for conceptually explaining the operation of the image signal collection device and method according to the present invention.

5 is a view schematically showing an image signal compression device according to another aspect of the present invention.

FIG. 6 is a diagram for describing an operation of the video signal compression device illustrated in FIG. 5.

7 is a flowchart illustrating a video signal compression method according to another aspect of the present invention.

8 is a diagram schematically illustrating an apparatus for decompressing a video signal according to another aspect of the present invention.

FIG. 9 is a diagram for describing an operation of an image signal decompression apparatus illustrated in FIG. 8.

10 is a flowchart illustrating a video signal decompression method according to another aspect of the present invention.

One aspect of the present invention for achieving the above object relates to an apparatus for collecting an image signal based on the importance determined according to the amount of information required by the user included in the image signal. An image acquisition apparatus according to an aspect of the present invention includes an image signal collector, an importance calculator, and an image processor. The image signal collector collects an image signal having a predetermined resolution at a predetermined frame rate, the importance calculator calculates the importance of the collected image signal, the image processor receives the image signal, and processes the image signal according to the importance, Generates collection information according to the performed image processing. In addition, when the importance level is less than the predetermined first threshold value, the image processing unit ignores the video signal and records the output in the collection information. Further, the image processing unit may reduce the resolution of the image signal when the importance is less than or equal to the second threshold value greater than the first threshold value, and record the effect in the collection information to output the image signal together with the processed image signal. . In addition, the image signal collector includes at least one or more cameras, and converts an output signal of any one of the cameras into a digital signal according to a channel selection signal and outputs the digital signal. Preferably, the importance calculator includes a previous frame storage unit that stores the image signal of the previous frame, and compares the received image signal with the image signal of the previous frame to calculate the importance according to the degree of change of the image signal. .

Another aspect of the present invention relates to an image signal collection method based on importance, which is determined according to an amount of information required by a user included in an image signal. The image signal collection method according to another aspect of the present invention collects an image signal. An image input step, a step of calculating the importance of the collected image signal, and an image processing the image signal according to the importance, and an image processing step of generating the collection information according to the performed image processing. Preferably, the image processing step according to another aspect of the present invention determines whether the calculated importance is less than the first threshold, and if so, ignores the collected image signal and records the output in the collection information to output the collected information. And further determine whether the calculated importance is smaller than the second threshold larger than the first threshold, and then reduce the resolution of the collected video signal and record the effect in the collection information to It is characterized by outputting together. In addition, the image input step is characterized in that the output signal of any one of the at least one camera selected by the channel selection signal is converted into a digital signal and collected as an image signal. More preferably, the importance calculation step may further include a previous frame storing step of storing the image signal of the previous frame, and comparing the received image signal with the image signal of the previous frame to calculate the importance according to the degree of change of the image signal. It is characterized by.

A third aspect of the present invention relates to an apparatus for compressing an image signal collected at a differential resolution, wherein the image signal compression apparatus according to the third aspect of the present invention includes a collection information reading unit, a central processing unit, and an image compression unit. do. The collection information reading unit receives and reads collection information including a resolution at which the received image signal is collected. The central processing unit receives the read-out collection information, sets a resolution to compress the video signal, and generates compression information including the set resolution information. In addition, the image compression unit is driven by the central processing unit, and compresses the received image signal to a set resolution and outputs the compressed image signal. Furthermore, the central processing unit may output the generated compressed information in synchronization with the compressed video signal. When there is no video signal corresponding to the collected information, the central processing unit does not drive the image compression unit. Characterized in that it is recorded on the compressed information and outputted. Preferably, the collection information includes at least one of importance and resolution information of the image signal. More preferably, the image compression unit may include a motion prediction processor and an image compression unit configured to perform motion prediction in response to a motion prediction command transmitted from an image compression processor that compresses an image signal received at a resolution set by the central processor. And at least one storage for storing the output signals of the processor and the motion prediction processor.

According to the fourth aspect of the present invention, a method for compressing an image signal collected at differential resolution includes receiving collection information including a resolution at which an image signal to be compressed is collected, and determining whether an image signal corresponding to the collection information is read. Step, when the video signal is read, setting the resolution to compress the video signal from the collection information, compressing the video signal with the set resolution, generating compressed information including the resolution and if the video signal is not read, Is generated as compressed information. Preferably, the collection information includes at least one of importance and resolution information of the image signal.

An apparatus for decompressing an image signal compressed at differential resolution according to the fifth aspect of the present invention includes a compression information reading unit, a central processing unit, and an image decompression unit. The compressed information reading unit reads the compressed information according to the compressed state of the input signal from the received input signal. The central processing unit sets the resolution to decompress based on the read compressed information, and drives the image decompression unit. The image decompression unit decompresses the compressed image signal among the input signals to the set resolution and outputs the decompressed image signal. Preferably, the central processing unit outputs the decompressed previous image signal as an output signal without driving the image decompression unit when there is no image signal to decompress by analyzing the compression information. More preferably, the compressed information includes resolution information of the video signal and whether there is a compressed video signal corresponding to the compressed information. The image decompression unit may include: an image decompression processor which decompresses an image signal compressed at a resolution set by the central processing unit, a motion compensation processor and an image decompression processor which perform motion compensation in response to a motion compensation command transmitted from the image decompression processor; And at least one storage for storing output signals of the motion compensation processor.

According to a sixth aspect of the present invention, there is provided a method for decompressing an image signal compressed at a differential resolution, the method comprising: reading compressed information according to a state in which an input signal is compressed from a received input signal, analyzing the compressed information; Determining whether a video signal to be decompressed exists, determining that a video signal to be decompressed exists, setting a resolution to decompress the video signal from the compressed information, and decompressing the video signal to the set resolution; If the video signal is not read, the decompressed previous video signal is output. Preferably, the compressed information includes resolution information of the compressed video signal, presence or absence of a video signal corresponding to the compressed information, and the like.

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings which illustrate preferred embodiments of the present invention and the contents described in the accompanying drawings.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. For each figure, like reference numerals denote like elements.

1 is a view showing a video signal collecting device according to the prior art.

The video signal collecting device 100 illustrated in FIG. 1 includes an image signal collecting unit 110, an analog / digital converting unit 120, a multiplexer 130, an image processing unit 140, an output unit 150, and an audio input unit. 160 and microphone 170. The image signal collector 110 includes first to fourth cameras 112, 114, 116, and 118.

Image signals are simultaneously collected by the first to fourth cameras 112, 114, 116, and 118. The collected video signal is converted into a digital signal by the analog / digital converter 120. The first to fourth cameras simultaneously receive different video signals and transmit the received video signals. The multiplexer 130 performs a task for selecting only one video signal from the transmitted signals. The multiplexer 130 may also be called a switch or quad. In addition, the microphone 170 collects a voice signal. The collected voice signal is converted into a digital signal through the voice input unit 160 and transmitted to the image processor 140. The image processor 140 receives an image signal and an audio signal, synchronizes the two signals, performs a predetermined image process, and provides the output signal to the output unit 150. Predetermined video processing refers to operations such as analyzing and emphasizing the received digital video signal. In the image signal collection apparatus 100 in which the image processor 140 is not provided, the output signal of the multiplexer 140 may be directly provided to the output unit 150 without performing image processing.

The output unit 150 outputs an image signal. The output of the video signal may be provided to an image compression unit (not shown) or the like, and may be compressed or transmitted. As described above, a compression algorithm such as MPEG may be used to compress the video signal.

However, the video signal has a characteristic that neighboring frames are similar. Nevertheless, the image signal collection device shown in FIG. 1 receives and provides an image signal at the same frame rate and the same resolution. Compressing video signals collected at the same frame rate and the same resolution generates a large amount of data.

In the case of a DVR, not all video signals inputted are necessary. For example, when storing a video signal after working hours, if there is no defense (ie, there is no intruder) in the video signal, the amount of data in the video signal can be further reduced by ignoring the video signal at that time. In addition, if there is little variation in the video signal (less movement), the video signal does not need to be observed in detail. Therefore, in this case, even if the video signal is collected by lowering the resolution of the collected video signal, there is no problem.

Accordingly, the present invention provides a technique for determining how useful the collected video signal is to a user and optimizing the amount of data of the collected video signal based on the determination result.

2 is a block diagram schematically illustrating an apparatus for collecting video signals according to an embodiment of the present invention.

The video signal collection device 200 illustrated in FIG. 2 includes an image signal collector 210, a multiplexer 220, an analog / digital converter 230, a synchronizer 240, a importance calculator 250, and an image processor. 260, a sensor input unit 280, and a sensor unit 290.

The image signal collector 210 includes first to fourth cameras 212, 214, 216, and 218, and image signals are simultaneously collected by the first to fourth cameras 212, 214, 216, and 218. . The collected image signal may be an image signal obtained by photographing a different area according to each camera. For example, assuming that the DVR is installed in the bank, the first camera 212 is a video of people entering and exiting the entrance of the bank, the second camera 214 is a video of people who pull out the number tag, the third camera 216 ) Captures an image of the people at the window, and the fourth camera 218 may photograph the employees at the back of the window to generate an image signal. Only one image signal may be sequentially selected by the multiplexer 220. The multiplexer 220 may select one video signal in response to the channel selection signal Ch, and the channel selection signal Ch may be a signal whose state is changed by the user. The selected video signal is converted into a digital video signal D_IMG by the analog / digital converter 230.

In addition, the sensor unit 290 generates outputs of various sensors. The sensors constituting the sensor unit provide information on various situations in addition to the image signal. For example, the sensor may include an infrared sensor S1, a door opening and closing sensor S2, a voice signal sensor S3, and the like. The sensor input unit 280 processes the generated sensor signals to generate a sensor signal SNS. The sensor signal SNS may be a signal obtained by serially arranging output signals of the plurality of sensors S1, S2, and S3. Alternatively, the signal may be one selected from among output signals of the plurality of sensors S1, S2, and S3 using a multiplexer (not shown).

The digital image signal D_IMG and the sensor signal SNS generated as described above are synchronized by the synchronization unit 240. The synchronization unit synchronizes the digital image signal D_IMG and the sensor signal SNS generated in real time, so that the sensor signal SNS is generated together with the image signal D_IMG generated at a certain time. do. For example, the synchronization of the video signal and the sensor signal is performed to distinguish the video signal D_IMG at the moment when the output occurs from the door open / close sensor S2 and the video signal when the output occurs to the infrared sensor S1. do. The output signal of the synchronizer 240 is provided to the image processor 260 and the importance calculator 250 as the image signal SYC.

The importance calculator 250 analyzes the received signal SYC and calculates the importance of the received image. Importance is a measure that is determined by how much information a given video signal contains to a user, and is used to reduce the amount of data in the video signal. That is, the importance calculator 250 analyzes the synchronized video signal SYC and determines the importance, thereby reducing the data amount of the video signal according to the importance.

There are many ways to determine importance. For example, the importance calculator 250 may analyze the histogram of the synchronized video signal SYC and determine that the case where a lot of red appears on the screen is important. Alternatively, by detecting the outline of the synchronized image signal SYC, it may be determined that a person is important. Alternatively, the importance calculator 250 may further include a previous frame storage 255 to store an image signal of a previous frame, and then, when the difference between the image signal of the previous frame and the current frame becomes a predetermined level or more, it is important. May be determined. Alternatively, it may be determined that the moving case is important, and it may be determined that the case where the input of the external sensor occurs is important. Furthermore, the importance calculator 250 may determine the importance by combining these various elements.

For example, if the importance is judged differently, in the case of DVRs installed in banks, it is recognized that the screen when customers do not come from the video after working hours is not important, but the screen after the door is opened and someone enters May be considered important. In addition, it may be judged that the appearance of a person wearing a mask on a face is much more important than when a person who does not wear a face appears. On the other hand, by applying a more advanced image processing technique, it is possible to recognize the face shape of the person in the image signal and compare it with the face of a particular person (criminal), it may be determined that the similar case is important.

In another case, consider a case of relaying a soccer game. In the case of a football game, a lot of lawns appear on the screen, so the color of the lawn does not have a big influence in determining the importance. In addition, it can be judged that when the soccer ball is near the goal line, it is much more important than when the soccer ball is near the center line. Therefore, the importance calculator 250 may determine that the image in which the goal post appears on the screen is more important than the importance of the image including the center line on the screen. Alternatively, the crowd may be used as a parameter to determine the importance. In this case, the importance calculator 250 may analyze the output of the voice sensor and determine the importance according to the amplitude of the output voice signal.

The importance calculator 250 may determine the importance by combining the outputs of various sensors. For example, if SO1 is the output of the infrared sensor (S1), SO2 is the output of the door opening and closing sensor (S2), and SO3 is a voice signal sensor (S3), the overall importance through the equation [1] (SGN) can be calculated.

Here, a1 to a3 are weights given to the respective sensor outputs. By setting the weight in various ways, the user can more precisely calculate the importance of the image required for the user. The importance SGN calculated by the importance calculator 250 is transmitted to the image processor 260.

The operation processor 260 receives the synchronized image signal SYC and the importance signal SGN and performs image processing on the synchronized image signal SYC according to the importance SGN. For example, the importance is calculated on a scale of 0, 1, 2, and the resolution of the received video signal is 640 * 480. Then, if the importance is 2, the video signal is the most important, so that the resolution is maintained and output to the output unit 270 as the image signal IMG that is processed. In addition, it records in the collection information CAP that the resolution of the video signal is maintained at 640 * 480, and outputs it to the output unit 270. Or, if the importance is 1, the video signal is less important, so the resolution is reduced to 320 * 240. Then, the image processor 260 outputs the collection information CAP including the image processed image signal IMG and information indicating that the resolution of the image signal is reduced, to the output unit 270. On the other hand, if the importance of the received video signal is determined to be zero, the video signal may be ignored. In this case, the resolution of the video signal is set to 0, the video signal of the frame is ignored, and the effect thereof is recorded in the collection information CAP and output to the output unit 270. Collection information (CAP) is information that stores the resolution of the video signal, the collected time, whether the various sensors are input. Once the transmitted video signal IMG is received, the collection information CAP can be analyzed and used to reproduce or compress the received video signal.

As described above, the image signal collection apparatus 200 according to an embodiment of the present invention calculates the importance SGN of the collected image signal SYC, and reduces the resolution of the image signal according to the calculated importance. The process is performed. This image processing occurs in a step before compressing the image, and serves to reduce the amount of data of the image signal. By reducing the video signal in the previous step of compressing the video signal, the hardware burden for transmitting and storing the video signal can be greatly reduced. In particular, in recent years, the need for transmitting and receiving video signals using a network increases, which has the advantage of lowering hardware specifications.

The configuration of the video signal collecting device 200 shown in FIG. 2 is not limited. For example, the sensor unit 290 and the image signal input unit 210 may operate with a greater number of sensors and cameras. In addition, the image processor 260 may not necessarily ignore the image signal when the importance SGN is the lowest, and may reduce the amount of data by image processing the image signal at the lowest resolution. However, if the video signal is ignored, the data amount can be further reduced. Therefore, it is interpreted that all techniques for differentially adjusting the data amount of the video signal according to the importance calculated for the video signal are included in the technical idea of the present invention.

3 is a flowchart illustrating a video signal collection method according to an embodiment of the present invention.

First, an image signal is input (S200). The video signal is input in real time by a plurality of cameras, and one of the video signals may be selected.

In addition, the sensor signal is input in synchronization with the video signal (S210). The sensor may include various sensors such as a door opening sensor, an infrared sensor, and a voice sensor, and a sensor signal combining the outputs of the sensors may be input.

Then, the importance of the video signal is calculated by comparing the input video signal with the video signal of the previous frame (S230). As described above, it is not necessary to compare the video signal of the previous frame with the received video signal to calculate the importance, and use only the difference value. The importance may also be calculated by analyzing the received video signal itself (e.g., when the histogram of a particular color is above a certain level, when there is an output of a door opener sensor, or when the output of a received voice sensor is above a certain level). .

When the importance is calculated, the received image signal is imaged using the calculated importance. For example, importance is calculated as 0, 1 and 2, and importance is 2 to maintain the resolution of the received video signal, 1 is to decrease the resolution, and importance is 0 to ignore the received video signal without transmitting it. Assume

Then, it is determined whether the importance is less than 2 (S240). If the importance is not less than 2, i.e., the importance is 2 in the illustrated case, the resolution is maintained without being reduced, and the image signal at which the resolution is maintained is stored and the effect is recorded in the collection information and transmitted (S250). The stored video signal can be used to calculate the difference value with the next video signal.

If it is determined in step S240 that the importance is less than 2, this time it is determined whether the importance is greater than zero (S260). If the importance is determined to be greater than zero, i.e., 1 in the example, the resolution of the received video signal is reduced. In addition, an image signal having a reduced resolution is stored, and the effect thereof is recorded and transmitted in the collection information (S270).

If it is determined in step S260 that the importance is not greater than zero, that is, 0, the received video signal is ignored, and the effect is recorded in the collection information and transmitted (S290).

When the image signal of the next frame is received, the above-described steps may be repeated to reduce the data amount by reducing the resolution of the image signal.

4 is a view for conceptually explaining the operation of the image signal collection device and method according to the present invention.

For convenience of explanation, it is assumed that the video signal is collected at a resolution of 640 * 480, and the importance of the collected video signal is calculated by being divided into 0, 1, and 2. 4 conceptually illustrates a result of image signals collected at the same resolution of 640 * 480 being image processed and transmitted by the image signal collection device and method according to an aspect of the present invention. For the sake of understanding, decreasing resolution is shown as a quadrangle square of the entire square.

The F1 frame determined to be 2 is transmitted and stored with the resolution of 640 * 480 maintained. Since the importance of the F2 frame is 1, the resolution is reduced to 320 * 240 and stored and transmitted. An importance of 1 means that the F2 frame is similar to the F1 frame or is not as important as the F1 frame, and thus the transmission is reduced regardless of the resolution. The F3 frame is the same as the F2 frame. Since the importance of the F4 frame is 0, the F4 frame is ignored and not stored or transmitted. That is, F4 to F7 frames are ignored. In this case, it means that the received video signal is exactly the same as the F3 frame or its importance is extremely low and may be ignored.

The importance of the F8 frame was determined to be 2. This means that an event has occurred, such as a sudden change in the screen that did not change or a door opening or closing signal. Therefore, the resolution of the received video signal is maintained at 640 * 480 and transmitted. Both F9 and F10 frames were determined to have importance 2. Therefore, since the screen of this section is an important screen, the resolution is maintained and transmitted. Since the importance of the F11 and F12 frames is determined as 1 again, the resolution is reduced, and the importance of the F13 frame is 2, so the resolution is maintained again. Finally, frames F14 through F15 are ignored.

The effect is recorded and transmitted in the collection information according to whether the resolution of each frame is maintained, reduced or the frame itself is ignored.

Since the image processing of FIG. 4 is by way of example, the technical concept of the present invention is not limited. 4 may be applied to all techniques for determining the importance of a received image signal and performing differential image processing according to the determined importance to reduce the amount of data. As described above, the technical idea of the present invention is to reduce the amount of data by performing image processing in a step before compressing a video signal. The video signal in which the data amount is reduced in this way can be reproduced correspondingly again by reading the collection information in which each frame is collected. That is, if each video signal is reproduced according to the collected resolution or the video signal of the corresponding frame is ignored, the first video signal may be displayed by outputting the video signal of the previous frame as it is. According to the present invention, the amount of data can be reduced without compressing the video signal.

5 is a view schematically showing an image signal compression device according to another aspect of the present invention.

According to another aspect of the present invention, an image signal compression apparatus 300 includes a collection information reader 310, a central processor 320, an image compressor 330, a compressed information generator 340, and an output 350. Include. In addition, the image compressor 330 includes an image compression processor 332, a motion prediction processor 334, and first and second storage units 336 and 338.

The collection information reader 310 reads out the collection information CAP of the image signal. The collected collection information CAP may include signals such as a time when the image signal is collected, a collected resolution, and a sensor input. The collection information CAP is transmitted to the central processing unit 320. The central processing unit 320 analyzes the collection information CAP to determine the resolution to compress.

The reason for determining the resolution to compress in the central processing unit 320 is as follows. That is, the image signals collected by the image signal collection device and method according to one aspect of the present invention are collected or ignored at different resolutions. Therefore, compression must be performed in accordance with the collected resolution.

As described above, a method of removing temporal redundancy between adjacent frames is used to compress an image signal. However, if the resolution of adjacent video signals is different, the temporal redundancy between the two frames is increased. Eliminating the increased temporal redundancy generates a large amount of data.

For example, assume a case of compressing the video signals shown in FIG.

F1 frames are compressed at a resolution of 640 * 480. F1 frames are compressed by removing their spatial redundancy, like I frames. In this case, DCT or wavelet transform may be used. In addition, the F1 frame is stored in the compression information to output that the compression is 640 * 480 resolution. Then, when decompression, the F1 frame is read out and the F1 frame is output at 640 * 480 resolution.

The resolution of the F2 and F3 frames was reduced to 320 * 240. When the F2 frame is compared with the F1 frame, the resolution of the F2 frame is reduced, so that the difference is increased when compared with the F1 frame. Therefore, if the F1 and F2 frames are compressed at the same resolution of 640 * 480, the amount of compressed data is also increased. Thus, F2 and F3 frames are compressed to 320 * 240, which is a reduced resolution. At this time, the collection information reading unit 310 reads out information on the resolution at which the F2 and F3 frames are collected from the collection information CAP. Then, the F2 frame is treated as an I frame at 320 * 240 resolution and compressed, and the F3 frame is compressed by removing temporal redundancy compared to the F2 frame. In addition, the F2 and F3 frames are stored in the compressed information that the compression is 320 * 240 and outputs. Then, when decompressing, the compressed information is read and the F2 and F3 frames are output at 320 * 240 resolution.

In the case of an F4 frame, the resolution is zero and the video signal is ignored. As described above, when the F4 frame is compressed by eliminating temporal redundancy compared to the F3 frame, a large amount of data is generated. Therefore, the F4 frame is regarded as the same as the F3 frame and is not compressed. The F4 frame is stored in the compressed information and output. Then, in the process of decompressing, the compressed information is read, and in the case of the F4 frame, it is not compressed and the F3 frame is output as it is. The F4 to F7 frames are cases in which the video signal is ignored because the importance of the video signal is 0. Accordingly, the F4 to F7 frames are not compressed. In addition, when decompression is performed, the effect is read, and the video signal of the F4 to F7 frames is replaced with the video signal of the F3 frame.

In the case of the F8 frame, since the data was collected at a resolution of 640 * 480, the effect is read from the collection information, and the F8 frame is compressed to 640 * 480. Since F9 and F10 frames were collected at a resolution of 640 * 480, they were compressed in a way that eliminates temporal redundancy with F7 frames.

As described above, the video signal compression apparatus according to another aspect of the present invention can significantly reduce the amount of data by compressing the video signal generated by the video signal collection apparatus according to one aspect of the present invention.

The configuration and operation of the image compression processor 332, the motion prediction processor 334, and the first and second storage units 336 and 338 are the same as those used in the conventional compression algorithm.

An example of using the MPEG compression algorithm is as follows.

5 and 6 will be described the operation of the video signal compression apparatus. The collection information reader 310 reads out only the collection information CAP from the received signal and transmits the collected information CAP to the central processing unit 320, and transmits the image signal IMG to the image compression unit 330. The collection information CAP includes information about a resolution at which an image signal is collected. The image compression processing unit 332 of the image compression unit 330 receives digital image data, stores it in the first storage unit 336, reads the stored data, and generates compressed image data, ie, I picture, which has been removed from spatial redundancy. do. In addition, the motion prediction request signal REQ_PRED is transmitted to the motion prediction processing unit 334 whenever necessary. The motion prediction processing unit 334 stores the compressed image (I picture) generated by the image compression processing unit 332 in the second storage unit 338, and predicts a future predicted encoded image (P picture) predicted from the compressed image data. Obtained and stored in the second storage unit 338 again. In addition, the motion prediction processing unit 334 obtains a bidirectionally encoded image (B picture) predicted from the I picture and the P picture, and stores the bidirectional encoded picture (B picture) in the second storage unit 338. That is, a motion vector is obtained using a future image signal value different from the compressed video signal, and a compressed predicted image is generated from the obtained vector. The motion prediction processing unit 334 transfers the result of performing the motion prediction from the I picture (END_PRED), that is, the P picture and the B picture to the image compression processing unit 332, and then predicts the motion-prepared data, that is, the I picture, to the next prediction. For reconstruction, the reconstructed data RECONST is retransmitted to the motion prediction processor 334. The image compression processor 332 transmits the bitstream data (RLC DATA) from which the run length code (RLC) encoding is completed from the I picture, the P picture, and the B picture to the central processing unit 320. The central processing unit 320 compresses the bit stream data (RLC DATA) one more time using the Huffman encoding method, adds start bits, commands and addresses, and synchronizes audio data with image data. Generate compressed data, that is, MPEG data. In addition, the compressed information generator 340 generates compressed information of the compressed MPEG data transmitted from the central processor 320. The compression information CMP includes information about a resolution at which the image signal is compressed. That is, the central processing unit 320 collects the collection information CAP, determines the resolution of the collected image signal, and compresses the compression rate control signal CMP_CON to the image compression processor 332 to perform compression according to the determined resolution. send. According to the received compression ratio control signal CMP_CON, the image compression processor 332 compresses the image signal.

A case of compressing image information of a frame such as the example shown in FIG. 4 will be described.

In the case of the F1 frame, since a video signal is collected at a resolution of 640 * 480, it is compressed at a resolution of 640 * 480. Accordingly, by analyzing the collection information CAP, the central processing unit 320 transmits a compression rate control signal CMP_CON to the image compression processing unit 332, which sets the resolution to 640 * 480, and thus the image compression processing unit 332 Compression is performed. In addition, the image compression processor F1 receives a picture type setting signal PIC_TYPE indicating that the F1 is to be compressed into an I picture type, and compresses the F1 frame into an I picture type.

F2 was collected at 320 * 240 resolution. Therefore, F2 is also collected at a new resolution, so it is compressed to an I picture type, and the resolution at that time is 320 * 240. Therefore, the central processing unit 320 transmits a compression ratio control signal CMP_CON and a signal PIC_TYPE for determining a picture type to the resolution of 320 * 240 to the image compression processing unit 332, and the image compression processing unit 332 accordingly. Compression is performed.

Since F3 is collected at the same resolution as F2, the central processing unit 320 transmits a signal CMP_CON to set the resolution of F3 to 320 * 240 and a signal PIC_TYPE to set the picture type to P type.

Since the video signals are not collected in the frames F4 to F7, the central processing unit 320 reads the collection information CAP to determine that the video signals have not been collected, and the image compression processor 332 compresses the frames. Inform that there is no need. Instead, the compressed image of the final frame F3 frame is regarded as the compressed image of the F4 to F7 frame as it is.

In the compression process as described above, the compression information generation unit 340 generates as the compression information (CMP) whether the resolution in which each frame is compressed and whether the frame is ignored. In the above example, the compression information (CMP) indicates that F1 is compressed at a resolution of 640 * 480, F2 and F3 are compressed at a resolution of 320 * 240, and image information of F4 to F7 is ignored, and instead of F3 Includes information to be replaced with video information.

7 is a flowchart illustrating a video signal compression method according to another aspect of the present invention.

First, the compressed video signal and the collection information are received (S710). The received collection information includes information on the resolution at which the image signal is received.

Then, it is compared whether the resolution information read out from the collection information is 0 (S720). If the collected resolution is determined to be 0, the video signal is ignored. Therefore, the video signal of the current frame is not compressed but is ignored, and only this is described in the compressed information (S730).

If the resolution is not 0, the resolution is compared with the resolution of the previous frame (S740). If the same, the received video signal is compressed to the existing resolution, and the effect is recorded in the compressed information (S760).

If the resolution is not the same as the resolution of the existing frame, the video signal is compressed to the new resolution and the effect is recorded in the compressed information (S750).

The video signal compressed at the existing resolution or the new resolution is stored (S770). A video signal that is not compressed and ignored is not stored. The reason is that the video signal of the frame is ignored. The stored video signal is used to remove temporal redundancy with the video signal of another frame.

The video signal compression method shown in FIG. 7 is illustrated only when the importance is calculated in three stages to collect the video signals, but the technical idea of the present invention is not limited thereto. Rather, it is interpreted that all techniques in which image signals are collected at different resolutions based on importance are included in the technical idea of the present invention.

8 is a diagram schematically illustrating an apparatus for decompressing a video signal according to another aspect of the present invention.

The video signal decompression device 800 according to another aspect of the present invention includes a compressed information reader 810, a central processor 820, a decompressor 830, and an output unit 850. In addition, the decompressor 830 includes an image decompression processor 832, a motion compensation processor 834, and first and second storage units 836 and 838.

The operation of the video signal compression apparatus will be described with reference to FIGS. 8 and 9.

The compressed information reader 810 reads only the compressed information CAP among the received signals and transmits the compressed information CAP to the central processing unit 820, and transmits the compressed image signal CMP_IMG to the image decompressor 830. Compression information (CMP) includes information about the resolution at which the image signal is compressed. The decompression processing unit 832 of the image decompression unit 830 receives the compressed image data, stores it in the first storage unit 836, and reads out the stored data to generate decompressed image data. In addition, whenever necessary, the motion compensation processing unit 834 transmits a motion compensation request signal REQ_COMP to perform motion compensation. The motion compensation processing unit 834 stores the decompressed image generated by the decompression processing unit 832 in the second storage unit 838, obtains an image obtained by performing motion compensation on the decompressed image data, and then reconstructs the second storage unit. Stored at 838. This decompression process is performed according to the picture type (PIC_TYPE) determined by the central processing unit 820. That is, in the case of an I picture, the decompression processing unit 832 decompresses the received video signal itself. In the case of a B picture and a P picture, the decompression processing unit 832 performs motion compensation from the video signal stored in the first storage unit 836 to perform compression. Release it.

The motion compensation processing unit 834 generates a result (END_COMP) of performing motion compensation from the I picture in the case of an image compressed with a P picture and a B picture as a type, and transmits the result (END_COMP) to the decompression processing unit 832. The reconstructed data RECONST is retransmitted to the motion compensation processor 834 again. The CPU 820 decodes the data encoded by the Huffman encoding method, and reads the start bit, the command, and the address to generate the RLC encoded bitstream data (RLC DATA). In addition, the decompression processing unit 832 receives and decodes the bitstream data RRL DATA from the central processing unit 820.

In this decompression process, the compressed information read by the compressed information reader 810 is used. That is, the central processing unit 820 receives the compression information CMP, determines the resolution of the compressed video signal, and decompresses the decompression control signal DECMP_CON to decompress according to the determined resolution. To send). In accordance with the received decompression control signal DECMP_CON, the decompression processing unit 832 compresses the video signal.

For convenience of explanation, a case of decompressing compressed image information obtained by compressing image information of a frame as shown in FIG. 4 will be described.

In the case of the F1 frame, since the video signal is compressed at the resolution of 640 * 480, it is decompressed at the resolution of 640 * 480. Therefore, the central processing unit 820 analyzes the compression information CMP and transmits the decompression control signal DECMP_CON to the decompression processing unit 832 that sets the resolution to 640 * 480, and accordingly the decompression processing unit 832. Performs compression. In addition, the decompression processing unit receives the picture type setting signal PIC_TYPE indicating that the F1 is compressed to the I picture type, and decompresses the F1 frame.

The F2 is compressed to 320 * 240 resolution. Therefore, the central processing unit 820 transmits a decompression control signal DECMP_CON for setting the resolution to 320 * 240 and a signal PIC_TYPE that determines the picture type of the compressed video signal to the decompression processing unit 832, and decompresses it. The processing unit 832 compresses accordingly.

Since F3 is collected at the same resolution as F2, the central processing unit 820 transmits a signal DECMP_CON to decompress the resolution of F3 to 320 * 240 and a signal that the picture type is P type (PIC_TYPE). Then, the decompression processing unit 832 decompresses F3 by generating a reconstruction signal RECONST motion-compensated by the motion compensation processing unit 834 in the stored previous frame F2.

Since the video signals are not compressed in the frames F4 to F7, the central processing unit 820 reads the compressed information (CMP) to determine that the video signal is compressed and has not been transmitted. Informs you that you do not need to unzip. Instead, the decompressed image of the final frame F3 is regarded as the image of the F4 to F7 frame as it is.

10 is a flowchart illustrating a video signal decompression method according to another aspect of the present invention.

First, a compressed video signal and compressed information are received (S910, S920). The compression information includes information on the resolution at which the video signal is compressed.

Then, it is compared whether a compressed video signal exists (S930). If no compressed video signal is present, it means that the video signal of that frame is not compressed. Therefore, the previous frame stored in place of the ignored video signal is output (S940).

If there is a compressed video signal, it is compared whether it is compressed at the same resolution as the decompression performed previously (S950). If it is determined that the video signal is compressed at the same resolution as that of the conventional decompression, the video signal is decompressed at the same resolution as the previously decompression (S960).

If the resolution of the received compressed video signal is not the same as the resolution of the previously decompressed video signal, the decompression step is performed at the resolution of the received compressed video signal (S970). The image signal decompressed to the existing resolution or the new resolution and the resolution of the decompression process performed are stored (S980). The stored video signal is then used when decompression at the same resolution or when no compressed video signal is received. When decompression is performed at the same resolution, it is used to remove temporal redundancy between the stored video signal and the received compressed video signal. Or, if there is no received compressed video signal, the stored video signal is output as it is.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible. For example, it is not necessary to perform image processing for reducing the resolution in an image signal evaluated as having low importance. Rather, it is also possible to perform image processing in which a particular display is made on a video signal of low importance. Therefore, the importance of the received video signal is evaluated, different image processing is performed according to the evaluated importance to collect the video signal, differentially compressing the collected video signal according to the importance, and the differentially compressed video signal All techniques for decompressing are included in the technical spirit of the present invention.

Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

According to the present invention, an apparatus and method for collecting image signals capable of reducing the amount of data of an image signal collected are provided.

In addition, the present invention provides an apparatus and method for compressing differentially collected video signals according to importance.

In addition, the present invention provides an apparatus and method capable of decompressing differentially compressed video signals.

Claims (25)

An apparatus for collecting a video signal based on the importance that is determined according to the degree to which the video signal is important to the user, An image signal collector configured to collect an image signal having a predetermined resolution at a predetermined frame rate; An importance calculator for calculating the importance of the collected video signals; And an image processor configured to receive the image signal, image the image signal according to the importance, and generate collection information according to the image processing performed. The method of claim 1, wherein the image processor, And if the importance is less than or equal to a first predetermined threshold, ignore the video signal and record the fact in the collection information to output the collected information. The method of claim 2, wherein the image processing unit, If the importance is less than or equal to the second threshold greater than the first threshold, And reducing the resolution of the image signal, recording the effect into the collection information, and outputting the image signal together with the image signal processed. The video signal collecting unit of claim 1, wherein the video signal collecting unit comprises: At least one or more cameras, And an output signal of one of the cameras is converted into a digital signal by a channel selection signal and output as the image signal. The method of claim 4, wherein the importance calculation unit, Including a previous frame storage unit for storing the video signal of the previous frame, And comparing the received image signal with the image signal of the previous frame to calculate the importance according to the degree of change of the image signal. The apparatus of claim 4, wherein the video signal collection device comprises: A sensor input unit configured to receive output signals of at least one or more sensors and generate a sensor output signal; And a synchronization unit for synchronizing the sensor output signal with the image signal to provide a synchronized image signal to the importance calculator and the image processor. The method of claim 6, wherein the importance calculation unit, And calculating the importance level by analyzing a sensor output signal of the synchronization unit. In the video signal collection method based on the importance that is determined according to the degree that the video signal is important to the user, An image input step of collecting an image signal; Calculating the importance of the collected video signal; and And an image processing step of processing the image signal according to the importance level and generating collection information according to the performed image processing. The method of claim 8, wherein the image processing step, And determining whether the calculated importance is smaller than a first predetermined threshold, and if so, disregarding the collected video signal and outputting the recorded information in the collection information. The method of claim 9, wherein the image processing step, It is determined whether the calculated importance is smaller than a second threshold larger than the first threshold, and then the resolution of the collected video signal is reduced and the effect is recorded in the collection information and output together with the video signal processed. An image signal collection method based on importance. The method according to any one of claims 8 to 10, wherein the image input step, An image signal collection method based on importance, wherein an output signal of one camera selected by a channel selection signal among at least one camera is converted into a digital signal and collected as the image signal. The method of claim 8 or 11, wherein the importance calculation step comprises: And a previous frame storing step of storing a video signal of a previous frame, And comparing the received video signal with the video signal of the previous frame and calculating the importance according to the degree of change of the video signal. The method of claim 8 or 12, wherein the video signal collection method, And synchronizing the output signal of at least one or more sensors with the collected video signal, And calculating the importance level by analyzing the synchronized image signal and output signals of the sensors. An apparatus for compressing an image signal collected at differential resolution, A collection information reading unit which receives and reads collection information indicating a state in which the received image signal is collected; A central processing unit which receives the collected information, sets a resolution to compress the video signal, and generates compression information including the set resolution information; A video compression unit driven by the central processing unit to output the compressed video signal by compressing the received video signal to the set resolution; And the central processing unit outputs the generated compressed information in synchronization with the compressed video signal and outputs the synchronized information. The method of claim 14, wherein the central processing unit, And if the video signal corresponding to the collection information does not exist, recording the effect into the compressed information without outputting the video compression unit and outputting the recorded information. The method of claim 14, wherein the collection information, And a video signal compression device having different resolutions. The method of claim 14 or 15, wherein the image compression unit, An image compression processor to compress an image signal received at a resolution set by the central processor; A motion prediction processor configured to perform motion prediction in response to the motion prediction command transmitted from the image compression processor; And at least one storage configured to store output signals of the image compression processor and the motion prediction processor. In the method for compressing a video signal collected at differential resolution, Receiving collection information indicating a state in which an image signal to be compressed has been collected; A determination step of determining whether an image signal corresponding to the collection information is read; When the video signal is read, setting a resolution to compress the video signal from the collection information, compressing the video signal with a set resolution, and generating compressed information including the resolution; and And if the video signal is not read, generating the message as the compression information. The method of claim 18, wherein the collection information, The video signal compression method of claim 1, wherein the video signal includes the resolution information collected. An apparatus for decompressing a video signal compressed at differential resolution, A compressed information reading unit which reads compressed information according to a state in which the input signal is compressed from the received input signal; A central processor configured to set a resolution to decompress based on the read compression information; And a video decompressor driven by the central processing unit and outputting a decompressed video signal by decompressing a compressed video signal of the input signal to the set resolution. Decompressor. The method of claim 20, wherein the central processing unit, When there is no video signal to decompress by analyzing the compression information, the decompressed previous video signal is output as an output signal without driving the video decompressing unit. Video signal decompression device. The method of claim 20, wherein the compressed information, And decompressing the video signal having the differential resolution, wherein the presence or absence of a compressed video signal corresponding to the resolution information of the video signal and the compressed information. The method of claim 20 or 21, wherein the image decompression unit, An image releasing processor for decompressing an image signal compressed at a resolution set by the central processing unit; A motion compensation processor configured to perform motion compensation in response to the motion compensation command transmitted from the image release processor; And at least one storage unit for storing output signals of the image decompression processor and the motion compensation processor. A method for decompressing a video signal compressed at differential resolution, Reading compressed information according to a state in which the input signal is compressed from a received input signal; Determining whether an image signal to be decompressed exists by analyzing the compressed information; If it is determined that there is a video signal to be decompressed, setting a resolution to decompress the video signal from the compression information, and decompressing the video signal at a set resolution; And if the video signal is not read, output the decompressed previous video signal. The method of claim 24, wherein the compressed information, And the resolution information of the compressed video signal and the presence or absence of a video signal corresponding to the compressed information.