JP2005236803A - Image compression device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To execute encoding processing by reading an image signal without a skip. <P>SOLUTION: The image compression device comprises a first storage part for storing the image signal, a writing part for writing the image signal in frame units into the first storage part, a frame number generating part, which generates a frame number corresponding to the image signal for every frame to be written by the writing part, a second storage part, which stores the frame number generated by the frame number generating part and has a first pointer for pointing the position storing the newest frame number and a second pointer for pointing the position storing a prescribed frame number, a reading part, which reads out the image signal corresponding to the frame number at the position pointed by the second pointer from the first storage part, and a control part which compares the position of the second pointer and that of the first pointer and controls the reading part so as to read out the image signal corresponding to the frame number at the position pointed by the second pointer from the first storage part in accordance with the comparison results. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image compression apparatus for moving images.

  JPEG-2000, the next-generation still image coding standard that has been standardized by the ISO / IEC (International Standardization Organization / International Electrotechnical Commission), in addition to its excellent compression performance, other areas in the image data can be used. Abundant functions such as ROI (Region Of Interest) function that encodes at a different compression rate than the other areas, progressive function that decodes while gradually improving image quality and resolution, and lossless compression function that does not cause information loss have. Widespread use is expected for next-generation digital still cameras, printers, and scanners. On the other hand, Motion JPEG-2000, which is a video version of JPEG-2000, inherits the excellent features of JPEG-2000 as it is and will be used for future camcorders, surveillance cameras, Internet video distribution, telemedicine distribution, video conferencing, etc. Is expected. Various techniques using the Motion JPEG-2000 have been proposed (see, for example, Patent Document 1).

  By the way, as shown in FIG. 6, the image compression apparatus adopting Motion JPEG-2000 is input to the image input unit 30 to which the image signal is input, the image buffer 31 for storing the image signal, and the image input unit 30. A writing unit 32 that writes image signals to the image buffer 31 in units of frames, a reading unit 33 that reads image signals stored in the image buffer 31, a control unit 34 that controls the reading unit 33, and a reading unit 33 An encoding unit 35 that encodes the image signal thus generated, a rate control unit 36 that generates an encoding amount determination signal that determines an optimal encoding amount based on the data encoded by the encoding unit 35, and a rate A packetizing unit 37 that packetizes the encoded data supplied via the control unit 36;

  The writing unit 32 writes the image signal supplied from the image input unit 30 to the image buffer 31 for each frame, and generates a signal indicating that the image signal has been written each time the image signal for one frame is written to the image buffer 31. It supplies to the control part 34.

  The reading unit 33 reads an image signal for one frame from the image buffer 31 and outputs the read image signal to the encoding unit 35.

  The encoding unit 35 is supplied with an image signal for one frame from the reading unit 33 and performs an encoding process on the supplied image signal. The encoding unit 35 supplies the image signal after the encoding process (hereinafter referred to as post-processing data) to the rate control unit 36. In addition, the encoding unit 35 generates a request signal for requesting the start and stop of the supply of the image signal, and supplies the request signal to the control unit 34.

  Based on the signal supplied from the writing unit 32, the control unit 34 increments a counter that counts the frame number by one. Further, the control unit 34 generates a predetermined control signal corresponding to the frame number indicated by the counter based on the request signal supplied from the encoding unit 35 and supplies the control signal to the reading unit 33.

  Based on the post-processing data supplied from the encoding unit 35, the rate control unit 36 generates a determination signal for determining the encoding amount in the encoding unit 35, and feeds back the generated determination signal to the encoding unit 35. Note that when the post-processing data supplied from the encoding unit 35 is larger than the predetermined data amount, the rate control unit 36 generates a determination signal for increasing the encoding amount, and the post-processing data is predetermined. If the amount of data is smaller than the data amount, a decision signal for reducing the amount of encoding is generated.

JP 2003-235042 A

  By the way, the rate control unit 36 takes a predetermined time to determine the optimum encoding amount, and the image buffer 31 is written regardless of the processing by the encoding unit 35 and the rate control unit 36. Image signals are sequentially written for each frame by the unit 32, and accordingly, a counter for displaying the frame number of the control unit 34 is also incremented sequentially.

  For example, the encoding unit 35 is supplied with the image signal corresponding to the frame number 1 from the image buffer 31, performs an encoding process on the supplied image signal, and then the image signal corresponding to the frame number 2 is the image buffer 31. When the encoding amount of the image signal corresponding to frame number 1 and frame number 2 is large when the supplied image signal is encoded, the time required for the encoding process is longer than usual. It will take.

  At this time, the image signal corresponding to the frame number 3 and the frame number 4 has already been written in the image buffer 31, and when the image signal corresponding to the frame number 5 is currently written, the frame number is displayed. In the counter, frame number 4 which is the latest frame number completely written in the image buffer 31 is displayed.

  Then, the control unit 34 causes the reading unit 33 to read out the image signal corresponding to the frame number 4 from the image buffer in response to a request signal for requesting the start of the supply of the image signal from the encoding unit 35. The image signal is supplied to the encoding unit 35. The encoding unit 35 encodes the image signal corresponding to the frame number 4 supplied from the reading unit.

  As described above, as a result of the time required for the encoding process by the encoding unit 35, the image signal corresponding to the frame number 3 is skipped (this is referred to as frame dropping).

  Therefore, in order to solve the above-described problem, the present invention requires time for encoding the image signal corresponding to the previous frame, and the image signal for the next frame even if the processing time of one frame is exceeded. An object of the present invention is to provide an image compression apparatus that does not drop frames.

  In order to solve the above problems, an image compression apparatus according to the present invention includes an image signal input unit to which an image signal is input, an encoding unit that encodes an image signal input from the image signal input unit, A control signal generation unit that generates a control signal for controlling the amount of encoding based on the data encoded by the encoding unit, and the encoding unit is configured to generate a predetermined signal based on the control signal generated by the control signal generation unit. In an image compression apparatus for encoding an image signal with an encoding amount, an image buffer for storing the image signal, a writing unit for writing the image signal input from the image signal input unit to the image buffer in units of frames, and a writing unit The frame number generating means for generating a frame number corresponding to the image signal for each frame written by, the frame number generated by the frame number generating means is stored, and the latest frame A queuing buffer having a first pointer that points to a location where a number is stored, a second pointer that points to a location where a predetermined frame number is stored, and a second pointer in the queuing buffer Reading means for reading out an image signal corresponding to the frame number of the position being read from the image buffer and supplying the read image signal to the encoding unit; and a position pointed to by a second pointer of the queuing buffer; The reading means is compared so that the position pointed to by the first pointer is compared, and the image signal corresponding to the frame number of the position pointed to by the second pointer is read from the image buffer according to the comparison result. Control means for controlling the second pointer of the queuing buffer to the first pointer. Controlled so as not to overtake the printer.

  An image compression apparatus according to the present invention includes an image buffer for storing an image signal, a writing unit for writing the input image signal to the image buffer in units of frames, a reading unit for reading the image signal stored in the image buffer, A control unit that controls the reading unit, a frame number generation unit that generates a frame number, and a queuing buffer that stores a frame number corresponding to each image signal in units of frames written in the image buffer, Control the coding index of the queuing buffer so that it does not overtake the image input index, and the buffer pointed to the coding index when a difference occurs between the image input index and the coding index. Read the corresponding image signal from the image buffer based on the stored frame number In, without skipping an image signal written in the image buffer can be supplied to the encoding unit.

  Hereinafter, an image compression apparatus for compressing a moving image will be described as an embodiment of the present invention.

  As shown in FIG. 1, the image compression apparatus 1 includes an image input unit 10 to which an image signal S1 is input, and a signal processing unit 11 that performs predetermined signal processing on the image signal S1 supplied from the image signal processing unit 10. An encoding unit 12 that performs an encoding process on a signal supplied from the signal processing unit 11; and rate control that generates a determination signal that determines an optimal encoding amount based on the data encoded by the encoding unit 12. Unit 13 and a packetizing unit 14 that packetizes the encoded data supplied via the rate control unit 13.

  The image input unit 10 receives the image signal S1 and supplies the image signal S1 to the signal processing unit 11. The image signal S1 is a stream signal, for example. The signal processing unit 11 performs predetermined signal processing on the image signal S <b> 1 supplied from the image input unit 10, and supplies the image signal S <b> 1 for one frame after processing to the encoding unit 12.

  The encoding unit 12 is supplied with an image signal S1 for one frame from a reading unit 22 of the signal processing unit 11 shown in FIG. 2 described later, and performs an encoding process on the supplied image signal S1. The encoding unit 12 supplies the image signal after the encoding process (hereinafter referred to as post-processing data) to the rate control unit 13. In addition, the encoding unit 12 generates a request signal for requesting start and stop of the supply of the image signal, and supplies the request signal to the control unit 23 of the signal processing unit 11.

  Based on the post-processing data supplied from the encoding unit 12, the rate control unit 13 generates a determination signal for determining the encoding amount in the encoding unit 12, and feeds back the generated determination signal to the encoding unit 12. The rate control unit 13 generates a determination signal for increasing the encoding amount 12 when the processed data supplied from the encoding unit 12 is larger than a predetermined data amount, and the processed data is When the amount of data is smaller than the predetermined amount of data, a determination signal for reducing the amount of encoding is generated.

  Here, a detailed configuration of the signal processing unit 11 will be described with reference to FIG. The signal processing unit 11 includes an image buffer 20 that stores an image signal, a writing unit 21 that writes the image signal input to the image input unit 10 to the image buffer 20 in units of frames, and an image signal stored in the image buffer 20. A reading unit 22 for reading out, a control unit 23 for controlling the reading unit 22, a frame number generating unit 24 for generating a frame number, and a frame number corresponding to each image signal written in the image buffer 20 in a frame unit. And a queuing buffer 25.

  The writing unit 21 converts the image signal S1 supplied from the image input unit 10 into an image signal for each frame and supplies the image signal to the image buffer 20. The writing unit 21 generates a writing completion signal S2 every time the writing of the image signal S1 for one frame to the image buffer 20 is completed, and supplies the writing completion signal S2 to the control unit 23. The image buffer 20 employs a so-called bank memory system in which a plurality of physical memories are switched and used in the same logical address space, and one frame of the image signal S1 is stored in each physical memory.

  The reading unit 22 reads a predetermined one frame of the image signal S1 from the image buffer 20 based on a signal supplied from the control unit 23 described later, and supplies the read one frame of the image signal S1 to the encoding unit 12. To do.

  Based on the write completion signal S2 supplied from the writing unit 21, the control unit 23 causes the frame number generating unit 24 that generates a frame number to generate a frame number and supplies the frame number to the queuing buffer 25.

  Further, the control unit 23 refers to the queuing buffer 25 according to the request signal supplied from the encoding unit 12, extracts a predetermined frame number, and supplies the frame number to the reading unit 22.

  As shown in FIG. 3, the queuing buffer 25 is composed of a ring buffer, and includes a first pointer P1 that points to a data writing position and a second pointer P2 that points to a data reading position. . The buffer stores a frame number corresponding to the image signal S1 written for each frame in the image buffer 20, and is controlled by the control unit 23 by a first pointer (hereinafter referred to as an image input index) P1. Is pointed to the buffer storing the latest frame number written based on the above, and the latest frame number read based on the control of the control unit 23 by the second pointer (hereinafter referred to as encoding index) P2. Point to the buffer where the is stored. The queuing buffer 25 is controlled so that the encoding index P2 does not overtake the image input index P1.

  Here, when the image signal S1 is written to the image buffer 20 for each frame by the signal processing unit 11, a frame number corresponding to the image signal S1 is generated and the frame number is written to the queuing buffer 25. A typical operation will be described with reference to the flowchart shown in FIG.

  In step S <b> 1, the control unit 23 determines whether or not the writing completion signal S <b> 2 is supplied from the writing unit 21. When the write completion signal S2 is supplied, the process proceeds to step S2.

  In step S2, the control unit 23 causes the frame number generation unit 24 to generate a frame number. The frame number generation unit 24 may be configured by a counter that counts frame numbers, and may be configured to increment the counter in accordance with control of the control unit 23.

  In step S <b> 3, the control unit 23 supplies the frame number generated by the frame number generation unit 24 to the queuing buffer 25.

  In step S4, the queuing buffer 25 stores the supplied frame number in the buffer. When a new frame number is stored in the buffer, the queuing buffer 25 moves (updates) the input image index P1 so as to point to the buffer. Each frame number stored in the queuing buffer 25 corresponds to a specific one-frame image signal stored in the image buffer 20.

  Next, the operation for extracting the image signal S1 stored in the image buffer 20 and supplying the extracted image signal S1 to the encoding unit 12 will be described with reference to the flowchart shown in FIG. Note that the control unit 23 performs an operation of extracting the image signal S1 from the image buffer 20 in accordance with the request signal supplied from the encoding unit 12.

  In step S10, the control unit 23 compares the image input index P1 and the encoding index P2 of the queuing buffer 25, and determines whether there is a difference. If there is a difference, the process proceeds to step S11. In the queuing buffer 25, the encoding index P2 is controlled so as not to overtake the image input index P1 as described above.

  In step S11, the control unit 23 moves the pointer of the encoding index P2 to the next buffer and reads the frame number stored in the buffer. The encoding index P2 points to the buffer from which the frame number was read during the previous frame number reading operation, but may point to the buffer from which the frame number is read during the next frame number reading operation. In this case, in step S11, the control unit 23 reads the frame number from the buffer pointed to by the coding index P2, and then moves the coding index P2 to the next buffer.

  In step S12, the control unit 23 controls the reading unit 22 so that the image signal S1 corresponding to the frame number read in step S11 is read from the image buffer 20. The reading unit 22 reads an image signal S1 for a predetermined frame from the image buffer 20 under the control of the control unit 23, and supplies the read image signal S1 to the encoding unit 12.

  Further, in order to associate the frame number with the image signal corresponding to the frame number, the control unit 23 reads the frame number generated by the frame number generation unit 24 in step S2, and writes the frame number to the writing unit 21. The frame number may be added to the image signal corresponding to the frame number written in the image buffer 20. In this case, the control unit 23 controls the reading unit 22 to read out from the image buffer 20 an image signal with a frame number that matches the frame number read from the queuing buffer 25 in step S12.

  In this way, in the image compression apparatus 1, the signal processing unit 11 that performs predetermined signal processing on the image signal S <b> 1 supplied from the image signal processing unit 10 includes the image buffer 20 that stores the image signal and the image input unit 10. A writing unit 21 that writes the input image signal to the image buffer 20 in units of frames, a reading unit 22 that reads the image signal stored in the image buffer 20, a control unit 23 that controls the reading unit 22, and a frame number A frame number generation unit 24 to be generated and a queuing buffer 25 that stores a frame number corresponding to each image signal written to the image buffer 20 in a frame unit, and an encoding index P2 of the queuing buffer 25 Is controlled so as not to overtake the image input index P1, and the image input index P1 and the encoding index. Based on the frame number stored in the buffer pointed to the encoding index P2 when a difference occurs between the encoding unit 12 and the encoding unit 12 Therefore, the image signal written in the image buffer 20 is not skipped.

  Further, the image compression apparatus 1 according to the present invention does not skip the image signal written in the image buffer 20 even when a low-speed processor is used, so that the accuracy of the encoded image is improved. Can do.

  Further, for example, the image compression apparatus according to the present invention can be applied to a recording system including an image compression apparatus such as a DVD (Digital Versatile Disk) recorder or an HDD (Hard Disk Drive) recorder.

It is a block diagram which shows the structure of the image compression apparatus which concerns on this invention. It is a block diagram which shows the structure of the signal processing part with which the image compression apparatus shown in FIG. 1 is equipped. It is a figure which shows the structure of the queuing buffer with which the signal processing part shown in FIG. 2 is equipped. 3 is a flowchart for explaining an operation of writing a frame number corresponding to an image signal written in an image buffer by the signal processing unit shown in FIG. 2 into a queuing buffer. 3 is a flowchart for explaining an operation when an image signal written in an image buffer is read by the signal processing unit shown in FIG. 2. It is a block diagram which shows the structure of the conventional image compression apparatus.

Explanation of symbols

  10 image input unit, 11 signal processing unit, 12 encoding unit, 13 rate control unit, 14 packetizing unit, 20 image buffer, 21 writing unit, 22 reading unit, 23 control unit, 24 frame number generation unit, 25 queue Ing buffer

Claims (2)

An image signal input unit that receives an image signal, an encoding unit that encodes the image signal input from the image signal input unit, and a control that controls the amount of encoding based on the data encoded by the encoding unit An image compression device that encodes an image signal with a predetermined coding amount based on the control signal generated by the control signal generation unit; and a control signal generation unit that generates a signal.
First storage means for storing image signals;
Writing means for writing the image signal input from the image signal input means to the first storage means in units of frames;
Frame number generating means for generating a frame number corresponding to an image signal for each frame written by the writing means;
The frame number generated by the frame number generating means is stored, the first pointer that points to the position where the latest frame number is stored, and the second that points to the position where the predetermined frame number is stored Second storage means having a pointer;
Reading out the image signal corresponding to the frame number at the position pointed to by the second pointer of the second storage means from the first storage means, and supplying the read image signal to the encoding unit Means,
The second pointer of the second storage means is compared with the position pointed to by the first pointer, and is pointed by the second pointer according to the comparison result. Control means for controlling the reading means so as to read an image signal corresponding to a frame number at a certain position from the first storage means,
The image compression apparatus according to claim 1, wherein the control means controls the second pointer of the second storage means so as not to pass the first pointer.   2. The image compression apparatus according to claim 1, wherein the second storage means is constituted by a ring buffer.

Images