JPH06141215A - Digital electronic still camera - Google Patents

Abstract

PURPOSE:To quickly perform compression processing and recording without using memory large in capacity in a continuous photographic mode. CONSTITUTION:When continuous photographing is started by turning on a continuous photographic switch 326, a compression circuit 324 performs the two-dimensional orthogonal transformation of image data image-picked up and accumulated in frame memory 320, and it is normalized and changed to a Huffman code, and it is recorded on a memory card 334 by controlling code quantity. At this time, a coefficient setting circuit 330 finds activity representing the characteristic of the image of first image data, and furthermore. finds a normalization coefficient for normalization and the coefficient of a bit distribution value for code quantity control based on the activity, and the coefficients calculated in the first image are set when the compression is performed on the images on and after second in the continuous photographic mode, which omits the computation of the coefficients in the operation of the image data on and after second. Thereby, compression coding time for one image for the images on and after second in the compression circuit 324 can be set at, for example, processing time with scanning time of nearly 30ms of one screen in an HTSC system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital electronic still camera, and more particularly to a digital electronic still camera capable of continuously shooting similar object images.

[0002]

2. Description of the Related Art In a digital digital electronic still camera, an image signal is converted into a digital signal and compressed into a data amount smaller than a desired data amount by a conversion method such as a two-dimensional orthogonal transformation, and a semiconductor memory, a magnetic disk or an optical disk. And the like on the image data recording medium. For example,
In the two-dimensional orthogonal transform method, one screen of image data is divided into blocks of a predetermined number and size, and the image data of each block is transformed into frequency domain data, that is, transform coefficients by two-dimensional orthogonal transform. A portion of the transform coefficient that is equal to or less than a predetermined threshold value is cut off, and the rest is divided by a quantization step based on the normalization coefficient, whereby normalization is performed. The normalized transform coefficients are then compressed by run-length and Huffman coding and stored in the image data storage medium. As a result, the storage capacity of the image data storage medium can be used efficiently. For details of the compression coding method such as the two-dimensional orthogonal transformation, refer to the pending patent application by the same applicant as this application, Japanese Patent Application No. 63-309870, and the like.

Further, in order to guarantee that a predetermined number of image data can be stored in an image data storage medium having a finite storage capacity such as a memory card, a pending patent application by the same applicant as this application, Japanese Patent Application No. -27499, by setting the bit allocation of each block in relation to the activity for each block for the image data of one screen, compressing in order from the lowest frequency component, and stopping the compression when this bit allocation value is reached, The amount of code after compression is constant.

In these compression coding methods, in order to perform predetermined coding and control of the code amount, a normalization coefficient at the time of normalization and a coefficient such as a bit distribution value are obtained for each image, and for each of these images. The calculated coefficient is used for compression coding for each image. That is, the image data of each captured frame is read out from the frame memory, the coefficient in each compression encoding is calculated based on the characteristic of the image and the encoded data amount, and the compression encoding is performed according to this coefficient. The code amount was controlled and recorded on the recording medium.

[0005]

However, in the conventional technique, various parameters required for compression are adaptively changed for each frame, that is, for each image, and when the fixed length is used, the parameters are changed. Since the amount of code is calculated, the parameters and the amount of code for each frame are set, and compression encoding is performed for each image, it takes a considerable time to compress and encode each image.
As a result, in the case of the continuous shooting mode as one mode of the camera, the compression encoding process cannot follow the reading of the image data from the image sensor, and a large-capacity memory capable of accumulating a plurality of uncompressed image data. It was difficult to practically take continuous shots without the provision.

The present invention solves the problems of the prior art as described above, and does not include a large capacity memory for accumulating a plurality of uncompressed image data in a digital electronic still camera using an orthogonal transform coding method. An object of the present invention is to provide a digital electronic still camera capable of realizing continuous shooting mode.

[0007]

In order to solve the above-mentioned problems, a digital electronic still camera according to the present invention has a continuous shooting mode for continuously shooting a scene image, and each scene image is captured. In a digital electronic still camera that converts an image signal representing the image data into a predetermined digital image data, compresses and encodes the image data and records the image data in a recording medium, the camera converts the image data representing each image into a predetermined size. An orthogonal transformation unit that divides into a plurality of blocks and orthogonally transforms them and outputs them in order from a DC component to a high-frequency component, and a normalization unit that normalizes each data orthogonally transformed by this orthogonal transformation unit. The normalization means for normalizing the image data based on the normalization coefficient set in accordance with the above, and the respective data normalized by this normalization means are provided. Coding means for coding by the coding method of (1), and code quantity control means for controlling the code quantity of the data for each block coded by this coding means based on the activity showing the characteristics of the image and outputting it to the recording medium. In the continuous shooting mode, this camera obtains the normalization coefficient and the code control amount based on the activity indicating the characteristics of the first image, and performs the compression coding of the first image data according to these coefficients. It is characterized in that the compression encoding processing of the image data of the second and subsequent images captured subsequently is performed based on the coefficients respectively obtained in the first image.

In this case, the digital electronic still camera is
A block activity calculating means for calculating a block activity indicating the characteristics of each block of the image supplied from the orthogonal transformation means, and a total activity calculating means for further calculating the total activity of the image from the information from the block activity calculating means. , A normalization coefficient setting means for obtaining a normalization coefficient based on the information from the total activity calculating means and the information in the normalization table and setting it in the normalizing means, and a calculated value from the block activity calculating means and the total activity calculating means Code amount setting means for calculating the bit distribution value in each block of the image data based on the above, and setting the bit distribution value of the image in the code amount control means, wherein the calculating means and the setting means are continuous shooting modes. At each of the first image frame The coefficient values are calculated and set in the normalizing means and the code amount controlling means, and the coefficients calculated in the first image are used in the normalizing means and the compressing encoding of the image data of the second and subsequent images, respectively. It may be set in the code amount control means.

The coefficients of the second and subsequent sheets may be set in the blanking period from the respective setting means to the normalizing means and the code amount control means.

[0010]

According to the digital electronic still camera of the present invention, when the first image is shot in the continuous shooting mode, the activity showing the characteristic of the image of this frame is obtained,
Based on this activity, a normalization coefficient for normalization and an appropriate code control amount for code amount control are calculated. The code control amount and the normalization coefficient are set in the normalization means and the code quantity control means. As a result, the orthogonally transformed first image data is normalized, coded, further subjected to code amount control, compression coded, and recorded on a recording medium. When the second image data is input during this period, the second image data is compression-encoded with the previous coefficient obtained for the first image and recorded on the recording medium. Similarly, if the image data for the third and subsequent images is also continuously input by continuous shooting, these image data will also be 1
The respective coefficients are compression-encoded with the coefficients obtained based on the image data of the first sheet and sequentially recorded on the recording medium.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a digital electronic still camera according to the present invention will be described in detail with reference to the accompanying drawings. Figure 1
Shows an embodiment of the digital electronic still camera according to the present invention. This digital electronic still camera converts the captured image of a subject into digital image data of the three primary colors of RGB, performs pre-processing such as white balance, and converts it to YCC image data of the NTSC system, and then converts it to a predetermined orthogonality. It is compression-encoded by the conversion encoding method and recorded on a recording medium such as a memory card. Particularly, in this embodiment, a digital electronic still camera having a continuous shooting mode in which continuous shooting is performed by pressing a specific button will be described.

In FIG. 1, a subject image passing through the lens 300 is a solid-state image pickup device 30 such as a CCD (Charge coupled device).
Imaged at 4. The lens 300 is position-controlled by a focusing mechanism 328 such as a contrast method or a phase difference method to focus an object image on the surface of the solid-state imaging device 300. On the surface of the solid-state imaging device 304, a color filter that transmits each color component of the RGB three primary colors for each pixel is mounted, and the color component that passes through is converted into an electric signal and output. Also,
In the continuous shooting mode, the solid-state imaging device 304 captures a continuous image by repeating electronic control, that is, accumulation and reading of charges exposed in each imaging cell by an electronic shutter.

The output of the solid-state image pickup device 304 is connected to an A / D (analog / digital) converter 308. The A / D converter 308 converts each pixel signal into 8-bit or 10-bit digital data and outputs it to the preprocessing circuit 312. The pre-processing circuit 312 corrects the white balance and gamma characteristics of the input image data and
This is a correction circuit that corrects the bias of the characteristic of the image data due to the bias of the characteristic of the filter or the like in the 300. The output of this preprocessing circuit 312 is connected to the process processing circuit 316.

The process processing circuit 316 is a conversion circuit for converting the input RGB image data into a luminance signal Y and color difference signals Cr and Cb for each pixel. The YCC image data converted by the process processing circuit 316 is stored in the frame memory 320, respectively. Frame memory 320
It is advantageous to store two pieces of image data, or to provide two frame memories each for one image.
In the case of this embodiment, a frame memory for one image may be used.
The image data stored in the frame memory 320 is read out by the compression circuit 324 and compression encoded.

The functional configuration of the compression circuit 324 in the present embodiment will be described with reference to FIG. 2. The circuit 324 includes a blocking section 10, a two-dimensional orthogonal transformation section 20, a normalization section 30, and an AC coding. It has a unit 40, a DC encoding unit 50, and a fixed length conversion unit 60. The blocking unit 10 includes a frame buffer, and the image data read from the frame memory 320 and stored in the frame buffer is stored in a plurality of blocks, for example, 8x.
Divides into blocks of 8 pixels and outputs.

The two-dimensional orthogonal transformation unit 20 performs two-dimensional orthogonal transformation on the image data of each block. As the two-dimensional orthogonal transform, known orthogonal transform such as discrete cosine transform and Hadamard transform is used. The image data for each block, which has been subjected to the two-dimensional orthogonal transformation, is arranged vertically and horizontally so as to be changed from low-order data to high-order data and is sequentially output. The DC component data is output first. That is, the image data as a result of the two-dimensional orthogonal transform, that is, the transform coefficient is sent to the normalization unit 30 in the order of the low frequency component of the alternating current component to the high frequency component in the unit of block.

The normalization unit 30 performs coefficient rounding down and normalization on this transform coefficient. Coefficient truncation is to compare the conversion coefficient with a predetermined threshold value and to discard the part below the threshold value. Further, the normalization unit 30 is a functional unit that performs quantization by dividing the truncated conversion coefficient by the quantization step value according to the normalization coefficient α. The normalization coefficient α is obtained based on the total value of the activities for each block, that is, the total activity as described later, and the normalization coefficient α is supplied from the coefficient setting unit 330 shown in FIG. Particularly in the case of the present embodiment, the normalization coefficient α is obtained based on the activity of the image data of the first image in the continuous shooting mode, and the normalization coefficient α of the first image is fixed for the second and subsequent images. Is used for.

The data normalized by the normalization unit 30 is scanned in a zigzag pattern in order from a low frequency to a high frequency AC component in block units, and is supplied to the AC encoding unit 40.
The AC encoding unit 40 encodes the transform coefficient input from the normalization unit 30. Since the AC component of the conversion coefficient is often continuous with zeros, a continuous amount of zero-value data, that is, a run length of zeros and a non-zero amplitude are obtained, and these are two-dimensionally Huffman-encoded. The image data for each block output from the AC encoding unit 40 is sent to the fixed length conversion unit 36. On the other hand, the DC component of the image data for each block output from the blocking unit 10 is sent to the DC encoding unit 50, and Huffman encoding is performed in the DC encoding unit 50. DC output from the DC encoding unit 50
The encoded data is sent to the fixed length conversion unit 60 before the AC component is output from the AC encoding unit 40.

The fixed length conversion unit 60 outputs the input Huffman-encoded variable length data into fixed length data of, for example, 8 bits. This fixed length part
60 is based on the bit allocation value β supplied from the coefficient setting unit 330 shown in FIG. 1, that is, the total data amount in each block constituted by the data of the DC component and the plurality of AC components is a predetermined data amount β. When the value exceeds the limit, it has a code amount control function of cutting off the subsequent data and controlling the code amount of each block within a predetermined range. The output of the fixed length conversion unit 60 is recorded on a recording medium such as the memory card 334 shown in FIG.

The coefficient setting section 330 obtains the bit distribution value β and the normalization coefficient α of each block based on the data for each block from the blocking section 10 or the two-dimensional orthogonal transformation section 20 of the compression circuit 324, and the compression circuit. This is the circuit set to 324. Particularly, in the present embodiment, when the continuous shooting switch 326 for setting the continuous shooting mode is turned on, the bit distribution value β and the normalization coefficient α in the first image data in continuous shooting are set.
Is also set in the same circuit 324 in the compression encoding of the second and subsequent sheets. The setting of coefficients for the second and subsequent sheets is performed during the blanking period. The normalization coefficient α obtained by the coefficient setting unit 330 is referred to from a look-up table (LUT) 332 stored in a memory such as a ROM (read only memory).

More specifically, as shown in FIG. 3, the coefficient setting section 330 includes a block activity calculating section 70, a total activity calculating section 80, a normalization coefficient setting section 90, and a bit allocation calculating section 100. The block activity calculation unit 70 calculates the activity for each block, that is, the degree to which the block includes image data of high frequency components. In the present embodiment, this block activity is obtained by adding the absolute values of the differences between the respective values of the pixel data forming one block and the average value of the pixel data of the block. In this embodiment,
This block activity is 1 in continuous shooting mode.
Only the block activity of the first image data is required.

The block activity output from the block activity calculating unit 70 is output to the total activity calculating unit 80 and the bit allocation calculating unit 100. The total activity calculating unit 80 adds the block activities for one screen to calculate the total value, that is, the total activity ACT, and outputs the total value to the normalization coefficient setting unit 90 and the bit allocation calculating unit 40. The normalization coefficient setting unit 90 sets the normalization coefficient α according to the total activity ACT to the compression circuit 32.
The normalization unit 30 of 4 is set. Particularly, in the continuous shooting mode, only the normalization coefficient α for the first image is obtained and set, and for the second and subsequent images, the normalization coefficient α obtained for the first image is set in the blanking period. The normalization unit 30 of the compression circuit 324 is set again. As the normalization coefficient α, a coefficient corresponding to the value ACT of the total activity of one screen is read from the look-up table 332 shown in FIG. 1, and in the present embodiment, this is corrected according to the number of blocks of image data. Then, the normalization coefficient α corresponding to the image is obtained.

The bit allocation calculating unit 80 calculates the bit allocation value β in each block from the ratio of the block activity to the total activity to determine the fixed length converting unit.
Output to 60. Particularly, in the present embodiment, the bit distribution value β is obtained from the block activity of the first image and its total activity in the continuous shooting mode, and when the first image is compressed and the second and subsequent images are compressed. The fixed length conversion unit 60 is commonly set. This setting is performed in each blanking period, similarly to the setting of the normalization coefficient α.

In the above structure, the continuous shooting switch
When 326 is turned on and a release button (not shown) is pressed, continuous shooting by the solid-state image sensor 304 is started. First, the image signal of the first subject image is converted into a digital signal by the A / D converter 308, and the preprocessing circuit 312
When each process is performed by the process circuit 316, the first image data is accumulated in the frame memory 320 as YCC image data. 1 in frame memory 320
When the YCC image data of the first sheet is accumulated, the compression circuit 324 is activated and the image data accumulated in the frame memory 320 is sequentially read by the blocking unit 10. From this, the coefficient acquisition shown in FIG. 4 and the compression processing of the first image are performed. This process will be described step by step. First, when the blocking unit 10 receives image data for one frame, it sequentially supplies this to the two-dimensional orthogonal transform unit 20 as block data of 8 × 8 pixels. Upon receiving this, the two-dimensional orthogonal transformation unit 20
The data of each block is two-dimensionally orthogonally transformed from a low-order component to a high-order component.

At this time, the block information of the blocked image data is sent to the coefficient setting unit 330, and the block activity in the first image is calculated by the block activity calculating unit 70. The activity for each block obtained by the block activity calculation unit 70 is sent to the total activity calculation unit 80 and the bit allocation calculation unit 100. Receiving this, the total activity calculating unit 80 calculates the total activity showing the characteristics of the first image and sends it to the normalization coefficient setting unit 90 and the bit allocation calculating unit 100. As a result, the normalization coefficient setting unit 90
Selects a normalization coefficient according to the total activity of the first image from the look-up table 322 and corrects it based on the number of blocks or the like to calculate a normalization coefficient α according to the image.

The calculated normalization coefficient α is stored in the compression circuit 324.
Is set in the normalization unit 30 of. Similarly, the bit allocation calculator that receives the block activity and the total activity
100 calculates the bit allocation value β in each block and sets it in the fixed length conversion unit 60. Almost 30 ms is spent for the coefficient calculation so far, and when this is completed, the compression is restarted with the set coefficient.

That is, the conversion coefficient of the AC component output from the two-dimensional orthogonal conversion unit 20 is compared with a predetermined threshold value, the portion below the threshold value is cut off, and the cut-off conversion coefficient is the normalization coefficient α. Quantization is performed by dividing by the quantization step value according to. Each of the normalized transform coefficients is Huffman coded by the AC coding unit 40 and sent to the fixed length conversion unit 60. On the other hand, the DC component of the image data blocked by the blocking unit 10 is sent to the DC encoding unit 50, Huffman-encoded, and supplied to the fixed length lengthening unit 60. The fixed length conversion unit 60, which has received the encoded data from both the encoding units 40 and 50, sequentially reduces the length of the low-order data of the DC component and the AC component and outputs the fixed length to the memory card 334. Bit allocation value β for which this output is set
When it exceeds, the fixed length conversion unit 60 cuts off the coded data that follows, waits for the data of the next block, and similarly outputs the fixed length. The output image data is
It is recorded in sequence on the memory card. It takes about 30 ms to compress and record the first image data.

When the compression encoding and recording of the first image data is completed, the compression circuit 324 sequentially reads out the second image data accumulated in the frame memory 320 to the blocking unit 10 and compresses it. Start encoding. In the case of the second image, the normalization coefficient α and the bit distribution value β obtained from the image data of the first image are set by the coefficient setting unit 330 during the blanking period. That is, in the coefficient setting unit 330, the normalization coefficient α previously obtained by the normalization coefficient setting unit 90 is set in the normalization unit 30 without referring to the second image data, and the bit allocation calculation unit The bit allocation value β obtained last time at 100 is set in the fixed length conversion unit 60. As a result, the second image data that has been blocked by the blocking unit 10 and orthogonally transformed by the two-dimensional orthogonal transformation unit 20 has the same coefficient as the previous normalization coefficient α in the normalization unit 10. Each is normalized and supplied to the AC encoding unit 40. On the other hand, the DC encoding unit 50 receives the DC component of the image data from the blocking unit 10, and Huffman-encodes the data, and the data has a fixed length before the AC component encoded data from the AC encoding unit 40. It is supplied to the conversion unit 60. The fixed length conversion unit 60, which has received the Huffman-encoded DC component and AC component, respectively, changes the fixed length in the same manner as the first frame, and the previous bit allocation value β
When it exceeds, the data after that is cut off and the code amount control is performed. Compressed recording of the second image data is approximately 30ms
It will be held at the time of.

Similar to the second image data, the image data of the third and subsequent images are not subjected to the calculation of the coefficient in the respective compression encoding, and the coefficient α of the compression encoding of the first image data is performed. , Β are set in the compression circuit 324 by the coefficient setting unit 330 during the blanking period, and the compression processing is performed based on these coefficients. As a result, the compression-coded third image data is sequentially recorded in the memory card 334 after the first and second image data.
The compression recording for the third and subsequent sheets is also performed in the period of 30 ms, respectively, similarly to the compression recording processing for the second sheet. This corresponds to a scanning speed of 1/30 sec for one screen in the NTSC system, for example, and therefore the continuous shooting speed can be almost achieved without depending on the compression coding speed. In this case, in continuous shooting, since the respective image data have almost the same image characteristics, that is, the activities, with respect to the same object scene image, the continuous shooting images having no practical problem are obtained. Can be obtained from the image data of.

Further, the frame memory 320 has a maximum capacity of two sheets for accumulating the first image data of the first image and accumulating the second image data in the coefficient calculation time of this image. Therefore, it is not necessary to mount a large-capacity frame memory for accumulating a plurality of image data to be continuously shot. At this time, since the first image data is read by the blocking unit 10 of the compression circuit 324, the writing of the second image data into the frame memory 320 and the writing of the first image data from the same memory 320 are performed. The first image data may be directly written to the frame buffer of the blocking unit 10 in the compression circuit 324 without passing through the frame memory 320, and the first image data may be read. The second image data captured during the compression process may be stored in the frame memory 320.

In the above embodiment, the case where the normalization coefficient α and the bit distribution value β are calculated and set in the compression circuit 324 has been described as an example, but the present invention is not limited to these coefficients. The setting of various parameters obtained based on the characteristics of each image in the orthogonal transform coding method is also included.

[0032]

As described in detail above, according to the digital electronic still camera of the present invention, in the continuous shooting mode, the coefficients such as the normalization coefficient and the code control amount calculated when the first image is compressed. Is used in the image compression of the second and subsequent images, it is possible to save the time for calculating the coefficients of the second and subsequent images. Therefore, it is possible to realize substantial continuous shooting with a camera using the orthogonal transformation method.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a digital electronic still camera according to the present invention.

FIG. 2 is a block diagram showing a functional configuration of a compression circuit in the embodiment.

FIG. 3 is a block diagram showing a functional configuration of a coefficient setting unit in the embodiment.

FIG. 4 is a diagram showing a time relationship of compression encoding in the continuous shooting mode of the embodiment.

[Explanation of symbols]

 10 Blocking unit 20 Two-dimensional orthogonal transformation unit 30 Normalization unit 40 AC coding unit 50 DC coding unit 60 Fixed length conversion unit 70 Block activity calculation unit 80 Total activity calculation unit 90 Normalization coefficient setting unit 100 Bit allocation calculation unit 324 Compression circuit 326 Continuous shooting switch 330 Coefficient setting section 332 Look-up table 334 Memory card

Claims (3)

[Claims] 1. A continuous shooting mode for continuously shooting an object scene image, converting image signals respectively representing the imaged object scene images into predetermined digital image data, and compressing the image data. In a digital electronic still camera which converts the data into a recording medium and divides the image data representing each image into a plurality of blocks each having a predetermined size and orthogonally transforms the image data, outputs the DC component to the high frequency component in order. Orthogonal transforming means and normalizing means for normalizing the respective data orthogonally transformed by the orthogonal transforming means, the normalizing means for normalizing the image data based on the normalizing coefficient set according to the characteristics of the image. An encoding means for encoding each data normalized by the normalizing means by a predetermined encoding method, and each block encoded by the encoding means A code amount control means for controlling a code amount of data based on an activity indicating an image characteristic and outputting the code medium to the recording medium, wherein the camera has an activity indicating a characteristic of a first image in a continuous shooting mode. The normalization coefficient and the code control amount are obtained based on the coefficient, the first image data is compression-encoded according to the coefficient, and the compression encoding processing of the second and subsequent image data captured is performed on the first image data. A digital electronic still camera, characterized in that processing is performed based on the coefficient obtained for each image. 2. The digital electronic still camera according to claim 1, wherein the camera calculates a block activity indicating a characteristic of each block of the image supplied from the orthogonal transforming unit, and a block activity calculating unit. Total activity calculating means for further calculating the total activity of the image from the information from the block activity calculating means, and the normalizing means for obtaining a normalization coefficient based on the information from the total activity calculating means and the information in the normalization table. The normalization coefficient setting means set to, and the bit distribution value in each block of the image data based on the calculated values from the block activity calculating means and the total activity calculating means, and the code amount control means Code amount setting to set bit allocation value The calculating means and the setting means calculate the coefficient values of the first image frame in the continuous shooting mode and set the coefficient values in the normalizing means and the code amount controlling means. A digital electronic still camera, wherein the coefficient calculated for the first image is set in the normalizing means and the code amount control means, respectively, when compressing and encoding the image data for the second and subsequent images. 3. The digital electronic still camera according to claim 2, wherein the coefficients of the second and subsequent images are set from the normalization coefficient setting means to the normalization means during the blanking period, and the bit allocation. A digital electronic still camera, wherein the code amount control unit is set by a calculating unit.