JP5103436B2 - Image processing apparatus, image processing method, and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program for performing image processing on an input image.

  The image captured by the sensor or camera includes various noises depending on the performance of the sensor or camera, the environment in which the image is captured, and the like. An image processing method for removing these noises and clearly distinguishing an object included in a captured image from a background has been reported. Hereinafter, an object included in a captured image is referred to as a target object.

  One of image processing methods for clearly distinguishing between a background and a target object included in a captured image is disclosed in Non-Patent Document 1, for example. In the method disclosed in Non-Patent Document 1, noise included in an image is removed by performing an erosion process and a dilation process, which are functions of morphology, on the image. . Shrinkage processing refers to a pixel product obtained by shifting the image in the horizontal and vertical directions in units of one pixel and the original pixel, and the logical product of density values between these pixels (in binary display). In this case, the logical product is calculated, and in the case of gradation display (shading display), the minimum value is calculated. In addition, the expansion processing is a logical sum of density values between these pixels by superimposing the original pixel and a pixel group obtained by shifting the image in the horizontal and vertical directions in units of one pixel (in binary display). In the case of logical sum, the maximum value in the case of gradation display (light and shade display)).

  In the method disclosed in Non-Patent Document 1, an image to be subjected to image processing is defined as X, a structural element is defined as B, (X + B) is defined as expansion of X by B, and (X−B) is defined as X by B. It is defined as contraction. In this case, the opening process (denoted as X ○ B) and the closing process (denoted as X · B), which are processes for removing noise, are respectively represented by the following expressions (1) and (2). Defined. The opening process is a process of performing an expansion process after performing a contraction process, and removes noise included in the background of the image. The closing process is a process of performing a contraction process after performing an expansion process, and removes noise (holes and fine depressions) included in the target object of the image.

X * B = (X-B) + B (1)
X · B = (X + B) −B (2)
Further, as a method specialized in the shape of the target object, for example, the Mehtre method, which is a registration image creation algorithm in fingerprint authentication, is disclosed in Non-Patent Document 2. In this method, a ridge direction of a fingerprint is obtained, and a Contextual filter is convolved along the direction to emphasize the ridge.

  In the two methods described above, single image processing is performed on input images of various image quality levels. For this reason, when the image quality level of the input image is remarkably low, noise remains in the output image that is a result of the image processing. That is, there is a problem that the image quality level of the output image is not sufficiently high.

  Thus, for example, Patent Document 1 discloses a method that can re-input an image when the image quality level of the output image is not sufficiently high. In the method disclosed in Japanese Patent Application Laid-Open No. 2004-228561, when fingerprint verification is performed, image processing (binarization, thinning, etc.) is performed on the user's fingerprint image read by the sensor. Then, the image quality level of the image subjected to the image processing is evaluated, and if it is determined that the image quality level is low, the user is notified of this. The user causes the sensor to read the fingerprint again according to the notification.

  In the technique disclosed in Patent Document 2, an input image is displayed on a monitor, and the user visually determines the image quality level of the displayed input image. When the user determines that the image quality level of the input image is low, the user performs image processing for correcting the deterioration of the image quality level. As a result, an image in which the deterioration of the image quality level is corrected is output.

JP-A-8-129644 JP 2000-324339 A

J. Serra, “Image Analysis and Mathematical Morphology,” Academic Press, London, 1982 B.M.Mehtre, “Fingerprint Image Analysis for Automatic Identification,” Machine Vision and Applications, vol.6, no.2-3, pp.124.139, 1993.

  If the techniques disclosed in Patent Document 1 and Patent Document 2 described above are used, more noise can be removed by repeating image input and correction even when the image quality level of the input image is extremely low. The image quality level of the output image can be increased.

  However, the technique disclosed in Patent Document 1 has a problem that it is necessary to re-input an image by the user, which places a burden on the user and requires time for re-inputting the image.

  In addition, the technique disclosed in Patent Document 2 requires correction of the image quality level by the user, requires training for the user, and requires time for correcting deterioration of the image quality level. .

  An object of the present invention is to provide an image processing apparatus, an image processing method, and a program that can efficiently obtain an output image with a high image quality level without imposing a burden on the user.

In order to achieve the above object, an image processing apparatus of the present invention receives an input of an image, generates a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image, An image processing apparatus that selects a processed image whose image quality level satisfies a predetermined condition from among the plurality of generated images,
An image input unit that receives an input of an image and outputs the received image as an input image;
Receiving an input image output from the image input unit, generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image, and generating the plurality of processed images An image processing unit for outputting an image;
An input image output from the image input unit is received, and the received input includes a plurality of determination image processes associated with each of the plurality of different types of image processing and having a smaller processing amount than the image processing. A determination image processing unit configured to generate a plurality of determination images associated with each of the plurality of processed images and to output the generated plurality of determination images;
The plurality of determination images output from the determination image processing unit are received as evaluation target images whose image quality levels are evaluated, the image quality levels of the received plurality of determination images are evaluated, and the plurality of determinations An image determination unit that outputs an evaluation result indicating a determination image whose image quality level satisfies a predetermined condition among the images for use;
A plurality of processed images output from the image processing unit and an evaluation result output from the image determination unit are received, and a determination image indicated by the received evaluation result among the plurality of received processed images An image selection unit for selecting a processed image corresponding to
A preprocessed image obtained by receiving an input image output from the image input unit and performing a preprocess for clarifying a difference in image quality level between the plurality of determination images with respect to the received input image. A pre-processing unit that generates and outputs the generated pre-processed image,
The determination image processing unit receives an input image output from the image input unit and a preprocessed image output from the preprocessing unit, and determines a predetermined one of the plurality of determination image processes. Performing image processing on the received input image, and performing image processing for determination other than the predetermined one image processing for determination among the plurality of image processing for determination on the received preprocessed image, A plurality of determination images are generated .

Further, the image processing method of the present invention includes:
Processing to accept image input ;
A process of generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image;
An image processing method in an image processing apparatus that executes a selection process for selecting a processed image whose image quality level satisfies a predetermined condition from among the plurality of generated images generated ,
The selection process includes
Each of the plurality of processed images is associated with each of the plurality of different types of image processing, and is subjected to a plurality of determination image processings having a smaller processing amount than the image processing. A determination image generation process for generating a plurality of determination images, which are evaluation target images that are associated and whose image quality level is evaluated;
A process of evaluating image quality levels of the plurality of generated determination images;
A process of selecting the processed image corresponding to a determination image whose image quality level satisfies a predetermined condition among the plurality of determination images;
Including
The determination image generation process includes:
Processing for generating a preprocessed image on which the preprocessing for clarifying the difference in image quality level between the plurality of determination images is performed on the input image;
A predetermined one of the plurality of determination image processes is applied to the input image, and a determination image process other than the predetermined one of the plurality of determination image processes is performed on the input image. Processing to generate the plurality of determination images by applying to a preprocessed image;
Including

The program of the present invention is stored in an image processing apparatus.
A procedure to accept image input ,
A procedure for generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image;
A selection procedure for selecting a processed image whose image quality level satisfies a predetermined condition among the plurality of generated images generated;
An image processing program for causing execution,
The selection procedure is:
Each of the plurality of processed images is associated with each of the plurality of different types of image processing, and is subjected to a plurality of determination image processings having a smaller processing amount than the image processing. A determination image generation procedure for generating a plurality of determination images, which are evaluation target images that are associated and whose image quality level is evaluated;
A procedure for evaluating image quality levels of the plurality of generated determination images;
A procedure for selecting the processed image corresponding to a determination image whose image quality level satisfies a predetermined condition among the plurality of determination images;
Including
The determination image generation procedure includes:
A procedure for generating a preprocessed image that has been subjected to preprocessing for clarifying the difference in image quality level between the plurality of determination images for the input image;
A predetermined one of the plurality of determination image processes is applied to the input image, and a determination image process other than the predetermined one of the plurality of determination image processes is performed on the input image. A procedure for generating the plurality of determination images by applying to a preprocessed image;
Including

  According to the present invention, the image processing apparatus receives an input of an image, and generates a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image. Then, a processed image whose image quality level satisfies a predetermined condition is selected from the generated processed images.

  Therefore, an output image with a high image quality level can be efficiently obtained without imposing a burden on the user.

1 is a block diagram showing a configuration of a first embodiment of an image processing apparatus of the present invention. It is a figure which shows an example of the input image which the image processing apparatus shown in FIG. 1 received, the 1st-3rd processed image, and the 1st-3rd image for a determination. It is a figure which shows the other example of the input image which the image processing apparatus shown in FIG. 1 received, the 1st-3rd processed image, and the 1st-3rd image for a determination. It is a figure which shows the other example of the input image which the image processing apparatus shown in FIG. 1 received, the 1st-3rd processed image, and the 1st-3rd image for a determination. FIG. 2 is a diagram for explaining an example of a method by which the image determination unit shown in FIG. 1 evaluates the image quality level of an image, where (a) shows an image to be evaluated, and (b) and (c) show the image quality level. It is a figure which shows the template image utilized for evaluation. 3 is a flowchart for explaining the operation of the image processing apparatus shown in FIG. 1. It is a block diagram which shows an example of the structure by which the 1st-3rd image processing part for a determination was abbreviate | omitted from the image processing apparatus shown in FIG. It is a figure which shows the other example of the input image which the image processing apparatus shown in FIG. 1 received, the 1st-3rd processed image, and the 1st-3rd image for a determination. FIG. 2 is a block diagram illustrating an example of a configuration in a case where an image processing unit performs image processing after an image selection unit receives an evaluation result in the image processing apparatus illustrated in FIG. 1. It is a block diagram which shows the structure of 2nd Embodiment of the image processing apparatus of this invention. The input image, the first preprocessed image and the second preprocessed image, the first to third processed images, and the first to third determination images received by the image processing apparatus shown in FIG. It is a figure which shows an example. 11 is a flowchart for explaining the operation of the image processing apparatus shown in FIG. 10. FIG. 11 is a block diagram illustrating an example of a configuration in a case where an image processing unit performs image processing after an image selection unit receives an evaluation result in the image processing apparatus illustrated in FIG. 10. It is a block diagram which shows the structure of 3rd Embodiment of the image processing apparatus of this invention. It is a block diagram which shows the structure of 4th Embodiment of the image processing apparatus of this invention. FIG. 16 is a block diagram illustrating an example of a configuration in which a threshold storage unit is provided in the image processing apparatus illustrated in FIG. 15. FIG. 16 is a block diagram illustrating an example of a configuration in which an image number storage unit is provided in the image processing apparatus illustrated in FIG. 15. FIG. 16 is a block diagram illustrating an example of a configuration in which an image storage unit is provided in the image processing apparatus illustrated in FIG. 15. FIG. 16 is a block diagram illustrating another example of a configuration in which an image storage unit is provided in the image processing apparatus illustrated in FIG. 15.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing the configuration of the first embodiment of the image processing apparatus of the present invention.

  As shown in FIG. 1, the image processing apparatus of the present embodiment includes an image input unit 10, a first image processing unit 20-1, a second image processing unit 20-2, and a third image processing unit 20. -3, first determination image processing unit 30-1, second determination image processing unit 30-2, third determination image processing unit 30-3, image selection unit 40, and image And a determination unit 50.

  The image input unit 10 receives input of an image to be subjected to image processing. The received images are output as input images to the first to third image processing units 20-1 to 20-3 and the first to third determination image processing units 30-1 to 30-3. .

  Each of the first to third image processing units 20-1 to 20-3 receives an input image output from the image input unit 10, and is a different type of image processing for the received input image. First to third processed images having undergone first to third image processing are generated. Then, the generated first to third processed images are output to the image selection unit 40. Details of the first to third image processing will be described later.

  Each of the first to third determination image processing units 30-1 to 30-3 receives an input image output from the image input unit 10, and performs different types of image processing on the received input image. First to third determination images subjected to certain first to third determination image processes are generated. Then, the generated first to third determination images are output to the image determination unit 50. Details of the first to third determination image processes will be described later.

  Here, the first to third image processing and the first to third determination image processing will be described.

  Here, first, a case will be described in which an input image randomly includes noise with a high density value (dark color) in the background and randomly includes noise with a low density value (light color) in the target object.

  FIG. 2 is a diagram illustrating an example of an input image, first to third processed images, and first to third determination images received by the image processing apparatus illustrated in FIG. 1.

  For example, the first image processing unit 20-1 performs a morphological smoothing process on the input image 11a as the first image process. The morphological smoothing process is a process applied to the input image in combination with the contraction process and the expansion process. In the morphological smoothing process, the process is performed in the order of shrinkage process, expansion process, expansion process, and shrinkage process.

  As a result of the first image processing, the first processed image 21a-1 from which the noise 1 having a high density value included in the background and the noise 2 having a low density value included in the target object are removed is obtained. Generated.

  For example, the second image processing unit 20-2 performs the second image processing that combines the smoothing processing and the contraction processing described above on the input image 11a. As a result of the second image processing, the second processed image 21a-2 from which the noise 1 having a high density value is more strongly removed is generated.

  For example, the third image processing unit 20-3 performs third image processing that combines the above-described smoothing processing and expansion processing on the input image 11a. As a result of the third image processing, a third processed image 21a-3 from which noise 2 having a low density value is more strongly removed is generated.

  The first to third processed images 21a-1 to 21a-3 are output to the image selection unit 40.

  Each of the first to third determination image processes is associated with each of the first to third image processes. That is, each of the first to third determination images 31a-1 to 31a-3 is associated with each of the first to third processed images 21a-1 to 21a-3. Furthermore, it is assumed here that each of the first to third determination image processes and each of the first to third image processes are the same process. That is, as shown in FIG. 2, the first determination image 31a-1 and the first processed image 21a-1 become the same image, and the second determination image 31a-2 and the second processed image 21a-2 becomes the same image, and the third determination image 31a-3 and the third processed image 21a-3 become the same image.

  The first to third determination images 31 a-1 to 31 a-3 are output to the image determination unit 50.

  Next, a case will be described in which the input image randomly includes a large number of high density noises in the background.

  FIG. 3 is a diagram illustrating another example of the input image, the first to third processed images, and the first to third determination images received by the image processing apparatus illustrated in FIG. 1.

  The first to third image processing units 20-1 to 20-3 perform the same image processing as described above on the input image 11b. That is, the first image processing unit 20-1 performs a smoothing process as the first image process on the input image 11b. The second image processing unit 20-2 performs second image processing that combines smoothing processing and contraction processing on the input image 11b. The third image processing unit 20-3 performs third image processing that combines smoothing processing and expansion processing on the input image.

  As a result, first to third processed images 21b-1 to 21b-3 as shown in FIG. 3 are generated, and the generated first to third processed images 21b-1 to 21b-3 are The image is output to the image selection unit 40.

  Also, the first to third determination image processing units 30-1 to 30-3 perform the first to third determination image processing on the input image 11b, thereby causing the first to first illustrated in FIG. Third determination images 31b-1 to 31b-3 are generated. Then, the generated first to third determination images 31 b-1 to 31 b-3 are output to the image determination unit 50. As described above, each of the first to third image processes and each of the first to third determination image processes are the same process.

  Next, a case where the input image randomly includes many low-density noises in the target object will be described.

  FIG. 4 is a diagram illustrating another example of the input image, the first to third processed images, and the first to third determination images received by the image processing apparatus illustrated in FIG. 1.

  The first to third image processing units 20-1 to 20-3 perform the same image processing as described above on the input image 11c. That is, the first image processing unit 20-1 performs a smoothing process as a first image process on the input image 11c. The second image processing unit 20-2 performs second image processing that combines smoothing processing and contraction processing on the input image 11c. The third image processing unit 20-3 performs third image processing that combines smoothing processing and expansion processing on the input image 11c.

  As a result, first to third processed images 21c-1 to 21c-3 as shown in FIG. 4 are generated, and the generated first to third processed images 21c-1 to 21c-3 are The image is output to the image selection unit 40.

  Further, the first to third determination image processing units 30-1 to 30-3 perform the first to third determination image processing on the input image 11c, so that the first to first illustrated in FIG. Third determination images 31c-1 to 31c-3 are generated. Then, the generated first to third determination images 31 c-1 to 31 c-3 are output to the image determination unit 50. As described above, each of the first to third image processes and each of the first to third determination image processes are the same process.

  Referring to FIG. 1 again, the image determination unit 50 uses the first to third determination images output from the first to third determination image processing units 30-1 to 30-3 as image quality. It accepts as an evaluation object image which evaluates the level. Then, any one of the first to third determination images is selected by evaluating the image quality levels of the received first to third determination images. Then, an evaluation result indicating any of the selected first to third determination images is output to the image selection unit 40.

  Here, an example of a method in which the image determination unit 50 evaluates the image quality levels of the first to third determination images will be described.

  FIG. 5 is a diagram for explaining an example of a method in which the image determination unit 50 shown in FIG. 1 evaluates the image quality level of an image, where (a) is a diagram showing an evaluation target image, and (b), (c) ) Is a diagram showing a template image used for evaluation of the image quality level.

  When evaluating the image quality of the image, the image determination unit 50 uses, for example, a template image as shown in FIGS. 5B and 5C for the noise included in the evaluation target image as shown in FIG. To detect by template matching. Then, the reciprocal of the total number of detected noises is used as an image quality score, and it is evaluated that the image quality is higher in descending order of image quality score.

  For example, in the example shown in FIG. 2, the image quality score increases in the order of the first determination image 31a-1, the second determination image 31a-2, and the third determination image 31a-3. Therefore, the image determination unit 50 outputs an evaluation result indicating the first determination image 31a-1 to the image selection unit 40. Similarly, in the example illustrated in FIG. 3, the image quality score increases in the order of the second determination image 31b-2, the first determination image 31b-1, and the third determination image 31b-3. . Therefore, the image determination unit 50 outputs an evaluation result indicating the second determination image 31b-2 to the image selection unit 40. Similarly, in the example illustrated in FIG. 4, the image quality score of the third determination image 31c-3 is the highest. Accordingly, the image determination unit 50 outputs an evaluation result indicating the third determination image 31c-3 to the image selection unit 40.

  Referring to FIG. 1 again, the image selection unit 40 includes first to third processed images output from the first to third image processing units 20-1 to 20-3, and an image determination unit. The evaluation result output from 50 is received. Then, a processed image corresponding to the determination image indicated by the received evaluation result is selected from the first to third processed images. Then, the image selection unit 40 outputs the selected processed image as an output image. Therefore, in the example illustrated in FIG. 2, the first processed image 21a-1 corresponding to the first determination image 31a-1 is an output image. Similarly, in the example illustrated in FIG. 3, the second processed image 21b-2 corresponding to the second determination image 31b-2 is an output image. In the example shown in FIG. 4, the third processed image 21c-3 corresponding to the third determination image 31c-3 is an output image.

  Hereinafter, the operation of the image processing apparatus configured as described above will be described. Here, description will be made based on the example shown in FIG.

  FIG. 6 is a flowchart for explaining the operation of the image processing apparatus shown in FIG.

  First, the image input unit 10 receives an input of an image (step S1).

  The image input unit 10 that has received the input of the image uses the received image as the input image 11a, and the first to third image processing units 20-1 to 20-3 and the first to third determination image processing units. 30-1 to 30-3.

  Each of the first to third image processing units 20-1 to 20-3 that has received the input image 11a output from the image input unit 10 is different from the first to third types of the received input image 11a. First to third processed images 21a-1 to 21a-3 subjected to the third image processing are generated (step S2). Here, the first image processing unit 20-1 performs smoothing processing as the first image processing on the received input image 11a. In addition, the second image processing unit 20-2 performs second image processing that combines smoothing processing and contraction processing on the received input image 11a. Further, the third image processing unit 20-3 performs third image processing that combines smoothing processing and expansion processing on the received input image 11a.

  Then, each of the first to third image processing units 20-1 to 20-3 outputs the generated first to third processed images 21 a-1 to 21 a-3 to the image selection unit 40.

  In addition, each of the first to third determination image processing units 30-1 to 30-3 that has received the input image output from the image input unit 10 is of a different type from the received input image 11a. First to third determination images 31a-1 to 31a-3 subjected to the first to third determination image processing are generated (step S3). As described above, each of the first to third determination image processes is the same as each of the first to third image processes.

  Next, each of the first to third determination image processing units 30-1 to 30-3 generates the generated first to third determination images 31 a-1 to 31 a-3 to the image determination unit 50. Output.

  The image determination unit 50 that has received the first to third determination images 31a-1 to 31a-3 output from the first to third determination image processing units 30-1 to 30-3, respectively, The image quality levels of the first to third determination images 31a-1 to 31a-3 are evaluated based on the image quality score (step S4). Then, the determination image with the highest image quality score is selected, and the evaluation result indicating the selected determination image is output to the image selection unit 40. Here, the image quality score increases in the order of the first determination image 31a-1, the second determination image 31a-2, and the third determination image 31a-3. Therefore, the image determination unit 50 outputs an evaluation result indicating the first determination image 31a-1 having the highest image quality score.

  The image selection unit 40 outputs the first to third processed images 21a-1 to 21a-3 output from the first to third image processing units 20-1 to 20-3 and the image determination unit 50. The evaluated result is received (step S5).

  Then, the image selection unit 40 selects a processed image corresponding to the determination image indicated by the received evaluation result (step S6), and outputs the selected processed image as an output image. Here, since the evaluation result indicates the first determination image 31a-1, the first processed image 21a-1 corresponding to the first determination image 31a-1 is output as an output image. .

  In the present embodiment, the processing performed by each of the first to third determination image processing units 30-1 to 30-3 and each of the first to third image processing units 20-1 to 20-3. Was the same. Thus, the first image processing unit 20-1, the first determination image processing unit 30-1, the second image processing unit 20-2, the second determination image processing unit 30-2, and the third The image processing unit 20-3 and the third determination image processing unit 30-3 can share a processor, and the image processing apparatus can be reduced in size and power consumption.

  In this case, the first to third determination image processing units 30-1 to 30-3 may be omitted. When the first to third determination image processing units 30-1 to 30-3 are omitted, the image determination unit 50 is output from the first to third image processing units 20-1 to 20-3. The first to third processed images are received as first to third determination images. Then, the image quality level of the received first to third determination images is evaluated.

  FIG. 7 is a block diagram illustrating an example of a configuration in which the first to third determination image processing units 30-1 to 30-3 are omitted from the image processing apparatus illustrated in FIG.

  The first to third image processing units 20-1 to 20-3 of the image processing apparatus illustrated in FIG. 7 send the first to third processed images not only to the image selection unit 40 but also to the image determination unit 50. Output.

  As described above, in the present embodiment, the image processing apparatus generates first to third processed images obtained by performing first to third image processing with different methods for removing noise on the received input image. To do. Then, a processed image with the least number of noises is selected from the generated first to third processed images.

  Therefore, an output image with a high image quality level can be efficiently obtained without imposing a burden on the user.

  In the present embodiment, the image determination unit 50 selects only the determination image having the highest image quality score from the first to third determination images. However, for example, the determination image having the second highest image quality score (or the subsequent order) may be selected. A plurality of determination images may be selected. This can be used when preparing a plurality of images of a high image quality level and a low image when creating a registered image for fingerprint authentication. For example, select a plurality of determination images with an odd or even number of image quality scores, select a predetermined number of determination images with the highest image quality score order, or select an image quality score in the middle of the image quality score order. A predetermined number of images for determination may be selected.

  Alternatively, a threshold value for the image quality score may be determined in advance, and the determination image may be selected by comparing the image quality score of the first to third determination images with the threshold value. For example, all the determination images having an image quality score equal to or higher than a predetermined threshold (or lower than the threshold) may be selected. Further, using all two predetermined threshold values (for example, T1 and T2 (T1 <T2)), all the determination images having an image quality score of T1 or more and less than T2 (or less than T1 and T2 or more) are selected. You may make it do.

  In addition, the first to third determination image processes can be simplified from the first to third image processes.

  FIG. 8 is a diagram illustrating another example of the input image, the first to third processed images, and the first to third determination images received by the image processing apparatus illustrated in FIG. 1.

  Here, the first determination image processing unit 30-1 performs an isolated point removal process with a smaller processing amount than the smoothing process on the input image 11d. The isolated point removal processing is, for example, when the density value of the focused pixel in the image is different from the density value of the 4 pixels adjacent to the focused pixel, the density value of the focused pixel is changed to the density value of the 4 adjacent pixels. Image processing. The first determination image 31d-1 that has been subjected to the isolated point removal process includes a noise 1 having a higher density value included in the background and a noise 2 having a lower density value in the target object than the input image 11d. Each is a reduced image.

  The second determination image processing unit 30-2 performs, for example, second determination image processing in which the above-described isolated point removal processing is combined with contraction processing. The second determination image 31d-2 subjected to the second determination image processing is an image from which the noise 1 having a high density value is more strongly removed.

  For example, the third determination image processing unit 30-3 performs third determination image processing in which the above-described isolated point removal processing is combined with expansion processing. The third determination image 31d-3 subjected to the third determination image processing is an image from which the noise 2 having a low density value is more strongly removed.

  The image determination unit 50 receives the first to third determination images 31d-1 to 31d-3 and determines the image quality levels of the received first to third determination images 31d-1 to 31d-3. When evaluating the image quality, for example, the template matching described above may be used. Here, the image quality score increases in the order of the first determination image 31d-1, the second determination image 31d-2, and the third determination image 31d-3, and the image determination unit 50 has the image quality score. An evaluation result indicating the highest first determination image 31 d-1 is output to the image selection unit 40.

  The image selection unit 40 outputs the first processed image 21d-1 corresponding to the first determination image 31d-1 indicated by the evaluation result output from the image determination unit 50 as an output image.

  In the image processing apparatus shown in FIG. 8, the first to third determination image processes are processes that are less accurate but have a smaller processing amount than the first to third image processes. Therefore, the evaluation result can be output to the image selection unit 40 without waiting for the first to third image processing in the first to third image processing units 20-1 to 20-3 to end. Therefore, the processing time can be shortened and the power can be reduced.

  Further, after the image selection unit receives the evaluation result output from the image determination unit, only the image processing unit that generates the processed image corresponding to the determination image indicated by the received evaluation result is caused to execute the image processing. It may be.

  FIG. 9 is a block diagram illustrating an example of a configuration in the case where the image processing unit performs image processing after the image selection unit receives an evaluation result in the image processing apparatus illustrated in FIG. 1.

  The image selection unit 40a illustrated in FIG. 9 executes the image processing only in any one of the first to third image processing units 20a-1 to 20a-3 after receiving the evaluation result output from the image determination unit 50. Command 201 is output. Thereby, low power can be realized. In the image processing apparatus shown in FIG. 9, it is not necessary to operate all of the first to third image processing units 20a-1 to 20a-3 at the same time. Therefore, the first to third image processing units 20a-1 to 20a-3 and the first to third determination image processing units 30-1 to 30-3 can share a processor, and the image processing apparatus Miniaturization and price reduction can be realized.

(Second Embodiment)
FIG. 10 is a block diagram showing the configuration of the second embodiment of the image processing apparatus of the present invention.

  The image processing apparatus according to the present embodiment includes a first preprocessing unit 60-1 and a second preprocessing unit 60-2 as compared with the image processing apparatus illustrated in FIG. 1 in the first embodiment. Is different. In FIG. 10, the number of preprocessing units is two, but the number of preprocessing units is not limited to two.

  The first preprocessing unit 60-1 receives the input image output from the image input unit 10, and generates a first preprocessed image subjected to the first preprocessing. Then, the generated first preprocessed image is output to the second determination image processing unit 30-2.

  The second preprocessing unit 60-2 receives the input image output from the image input unit 10, and has undergone a second preprocess that has been subjected to a second preprocess that is different from the first preprocess described above. Generate an image. Then, the generated second preprocessed image is output to the third determination image processing unit 30-3.

  Here, the first preprocessing, the second preprocessing, the first to third image processing, and the first to third determination image processing will be described. Here, a case will be described in which the input image randomly includes a large number of high density noises in the background.

  11 shows the input image, the first preprocessed image and the second preprocessed image, the first to third processed images, the first to third processed images received by the image processing apparatus shown in FIG. It is a figure which shows an example of the image for determination.

  The first preprocessing unit 60-1 receives the input image 11e output from the image input unit 10. And the 1st pre-processed image 61-1 which performed the 1st pre-process on the received input image 11e is produced | generated. Here, as an example, the first pre-processing is processing for randomly adding noise 2 having a low density value to the entire surface of the input image 11e. The generated first preprocessed image 61-1 includes some noises 2 with low density values in the target object, and some of noises 1 with high density values that were included in the background of the input image 11e. Becomes an image from which is removed. The first preprocessed image 61-1 is output to the second determination image processing unit 30-2.

  The second preprocessing unit 60-2 receives the input image 11e output from the image input unit 10. And the 2nd pre-processed image 61-2 which performed 2nd pre-processing on the received input image 11e is produced | generated. Here, as an example, the second pre-processing is processing for randomly adding noise 1 having a high density value to the entire surface of the input image 11e. The generated second preprocessed image 61-2 is an image in which the number of noises 1 having a high density value included in the background is greater than that of the input image 11e. The second preprocessed image 61-2 is output to the third determination image processing unit 30-3.

  The first to third determination image processing units 30-1 to 30-3 are the same determination images as the first to third determination image processing units 30-1 to 30-3 in the first embodiment. Process. As a result, first to third determination images 31e-1 to 31e-3 shown in FIG. 11 are generated, and the generated first to third determination images 31e-1 to 31e-3 are image determinations. Is output to the unit 50. Here, the input image 11e includes a lot of noise 1 having a high density value. Therefore, the image quality level of the third determination image 31e-3 to which the noise 1 having a high density value is added by the second preprocessing is the first determination image 31e-1 and the second determination image 31e-. 2 is significantly lower than the image quality level of 2.

  Note that the pre-processing is not limited to the above-described processing, and may be processing in which one or a plurality of morphological contraction processing, expansion processing, smoothing processing, and the like are combined.

  Hereinafter, the operation of the image processing apparatus configured as described above will be described. Here, description will be given based on the example shown in FIG.

  FIG. 12 is a flowchart for explaining the operation of the image processing apparatus shown in FIG.

  First, the image input unit 10 receives an input of an image (step S21).

  The image input unit 10 that has received the input of the image uses the received image as the input image 11e, and the first preprocessing unit 60-1, the second preprocessing unit 60-2, and the first determination image processing. Output to the unit 30-1 and the first to third image processing units 20-1 to 20-3.

  Each of the first to third image processing units 20-1 to 20-3 that has received the input image 11e output from the image input unit 10 performs the first to third image processing on the received input image 11e. The first to third processed images 21e-1 to 21e-3 are generated (step S22). Here, the first image processing unit 20-1 performs a smoothing process as a first image process on the received input image 11e. In addition, the second image processing unit 20-2 performs second image processing that combines smoothing processing and contraction processing on the received input image 11e. Further, the third image processing unit 20-3 performs third image processing that combines smoothing processing and expansion processing on the received input image 11e.

  Then, each of the first to third image processing units 20-1 to 20-3 outputs the generated first to third processed images 21e-1 to 21e-3 to the image selection unit 40.

  The first preprocessing unit 60-1 that has received the input image 11e output from the image input unit 10 performs the first preprocessed image 61- on which the first preprocessing is performed on the received input image 11e. 1 is generated. Similarly, the second preprocessing unit 60-2 that has received the input image 11e output from the image input unit 10 performs the second preprocessed processing on which the second preprocessing has been performed on the received input image 11e. An image 61-2 is generated (step S23). Here, the first preprocessing unit 60-1 performs first preprocessing for randomly adding noise 2 having a low density value to the entire surface of the input image 11e. In addition, the second preprocessing unit 60-2 performs a second preprocessing for randomly adding noise 1 having a high density value to the entire surface of the input image 11e.

  Then, the first preprocessing unit 60-1 outputs the generated first preprocessed image 61-1 to the second determination image processing unit 30-2, and the second preprocessing unit 60- 2 outputs the generated second preprocessed image 61-2 to the third determination image processing unit 30-3.

  The first determination image processing unit 30-1 that has received the input image 11e output from the image input unit 10, and the first preprocessed image 61-1 output from the first preprocessing unit 60-1. The second image processing unit 30-2 for determination and the third image processing unit for determination 3 that received the second preprocessed image 61-2 output from the second preprocessing unit 60-2 Each of 30-3 performs first to third determination image processing and generates first to third determination images 31e-1 to 31e-3 (step S24). Each of the first to third determination image processes may be the same process as the first to third image processes, or a process having a smaller processing amount than the first to third image processes as described above. But you can.

  Next, each of the first to third determination image processing units 30-1 to 30-3 supplies the generated first to third determination images 31e-1 to 31e-3 to the image determination unit 50. Output.

  Steps S25 to S27, which are subsequent operations, are the same as steps S4 to S6 in the operation flow shown in FIG. 6 in the first embodiment described above, and thus description thereof is omitted here. However, since the image quality score of the second determination image 31e-2 is the highest here, the second processed image 21e-2 corresponding to the second determination image 31e-2 is output as an output image. The

  As described above, in the present embodiment, the image processing apparatus is configured to clarify the difference in image quality level between the first to third determination images before performing the first to third determination image processing. Processing is performed on the input image. Therefore, it is possible to appropriately evaluate the image quality level of the image for determination even with an evaluation method with a small amount of processing and low accuracy, and it is possible to shorten the processing time for evaluating the image quality level and downsize the image processing apparatus. be able to.

  In this embodiment as well, in the same way as in the first embodiment described above, after the image selection unit receives the evaluation result output from the image determination unit, the processing corresponding to the determination image indicated by the received evaluation result is completed. Only the image processing unit that generates the image may be caused to execute the image processing.

  FIG. 13 is a block diagram illustrating an example of a configuration in the case where the image processing unit performs image processing after the image selection unit receives an evaluation result in the image processing apparatus illustrated in FIG. 10.

  The image selection unit 40a illustrated in FIG. 13 executes the image processing only in any one of the first to third image processing units 20a-1 to 20a-3 after receiving the evaluation result output from the image determination unit 50. Command 201 is output. Thereby, low power can be realized. In the image processing apparatus shown in FIG. 13, it is not necessary to operate all of the first to third image processing units 20a-1 to 20a-3 at the same time. Therefore, the first to third image processing units 20a-1 to 20a-3 and the first to third determination image processing units 30-1 to 30-3 can share a processor, and the image processing apparatus Miniaturization and price reduction can be realized.

  In the first embodiment and the second embodiment described above, the image determination unit and the image selection unit may receive the input image output from the image input unit 10 and add the received input image to the evaluation target image. Good. In this case, if the image selection unit determines that the image quality of the input image output from the image input unit 10 is sufficiently high, before receiving the first to third processed images output from the image processing unit, An input image can also be output as an output image. In this case, the processing time for performing image processing can be shortened, and low power can be realized. If the image determination unit 50 determines that the image quality of the received input image is sufficiently high, an evaluation result indicating that any of the first to third processed images may be used as the output image is displayed in the image selection unit. You may make it output. Thereby, the image selection unit can output the processed image having the shortest processing time among the first to third image processings as an output image, and can shorten the processing time for performing the image processing. , Low power can be realized.

(Third embodiment)
FIG. 14 is a block diagram showing the configuration of the third embodiment of the image processing apparatus of the present invention.

  Compared with the image processing apparatus shown in FIG. 1 in the first embodiment described above, the image processing apparatus according to this embodiment omits the first to third determination image processing units 30-1 to 30-3. Is different. Therefore, the image determination unit 150 directly receives the input image output from the image input unit 10 and evaluates the image quality level of the received input image.

  The image determination unit 150 receives the input image output from the image input unit 10 and detects noise included in the received input image by the template matching described above. Then, the image quality of the input image is evaluated based on the number of detected noises. Specifically, for example, the image determination unit 150 detects the number of high density noises (Nb) detected by the template shown in FIG. 5B and the template shown in FIG. The number of noises having a low density value (Nw) is detected separately. When Nb is sufficiently larger than Nw (Nb−Nw> C; C is an arbitrary value), the second processed image subjected to the second image processing in which noise with a high density value is more strongly removed It is determined that the image quality level is the highest. On the other hand, when Nb is sufficiently smaller than Nw (Nw−Nb> C; C is an arbitrary value), the third processed image subjected to the third image processing in which noise having a low density value is more strongly removed is applied. It is determined that the image quality level is the highest. In other cases (| Nw−Nb | <C), it is determined that the image quality level of the first processed image subjected to the first image processing is the highest. Then, the image determination unit 150 outputs an evaluation result indicating the determined processed image to the image selection unit 40.

  The image selection unit 40 receives the evaluation result output from the image determination unit 150. And the processed image which the received evaluation result shows among the 1st-3rd processed images is output as an output image.

  As described above, in the present embodiment, the image determination unit 150 has only one image for evaluating the image quality level. Therefore, the processing time can be shortened compared to the first embodiment and the second embodiment described above, and the power consumption and size of the image processing apparatus can be reduced.

  In this embodiment as well, as in the first and second embodiments described above, the image selection unit receives the evaluation result output from the image determination unit, and then the processed result indicated by the received evaluation result is displayed. Only the image processing unit that generates the image may be caused to execute the image processing. Thereby, low power can be realized.

(Fourth embodiment)
FIG. 15 is a block diagram showing the configuration of the fourth embodiment of the image processing apparatus of the present invention.

  The image processing apparatus according to this embodiment is different from the image processing apparatus illustrated in FIG. 1 in the first embodiment described above in that the image determination unit 250 outputs the re-input command 101 to the image input unit 210. Different. The re-input command 101 is a command for causing the image input unit 210 to accept an image again according to the result of the image quality level evaluation performed by the image determination unit 250.

  The image determination unit 250 receives the first to third determination images output from each of the first to third determination image processing units 30-1 to 30-3 and receives the received first to third images. The image quality level of each determination image is evaluated based on the image quality score of the determination image. Here, the image quality scores of the first to third determination images are Iq1 to Iq3, respectively, and the image quality scores are higher in the order of Iq3, Iq2, and Iq1. At this time, when the difference between the image quality scores is small (Iq3-Iq1 <C, C is an arbitrary constant), the image determination unit 250 outputs the re-input command 101 to the image input unit 210. On the other hand, when the difference between the image quality scores is large (Iq3-Iq1> C), the image determination unit 250 outputs an evaluation result indicating the third determination image having the highest image quality score to the image selection unit 40.

  The image input unit 210 that has received the re-input command 101 output from the image determination unit 250 receives the input of the image again, and outputs the received image as an input image.

  This is because when the image quality level of the input image is extremely low, the image quality levels of both the first to third processed images and the first to third determination images are low, and as a result, the difference between the image quality scores is reduced. It is utilized that it becomes smaller (Iq3-Iq1 <C).

  The condition for determining whether the image determination unit 250 outputs the re-input command 101 to the image input unit 10 or the evaluation result to the image selection unit 40 is arbitrarily set other than the above-described conditions. can do. For example, assuming that the image quality level of the input image is always greater than or equal to a certain level, if the difference between the image quality scores of the first to third determination images is large (Iq3-Iq1> C), the image input unit The re-input command 101 may be output to 210. In this case, if the difference between the image quality scores of the first to third determination images is small (Iq3-Iq1 <C), the evaluation result may be output to the image selection unit 40.

  As described above, in the present embodiment, when the difference between the image quality levels of the first to third determination images received by the image determination unit 250 is small, that is, any image quality level of the first to third determination images. Is lower, the image input unit 210 receives the input of the image again. Therefore, an image having a higher image quality level than the output image in the first to third embodiments can be obtained as the output image.

  Generally, in a fingerprint authentication device, an image with a different image quality level is input every time sensing is performed depending on the state of sweat on the surface of the finger. Similarly, even when shooting an object, the image quality level of the input image changes due to changes in ambient brightness and the like. In addition, the image may be re-input by intentionally changing the ambient brightness, the direction of the target object, or the parameters of the camera or sensor.

  The image determination unit 250 may also receive an input image output from the image input unit 210, and the received input image may be an evaluation target image together with the first to third determination images. In this case, the image quality score (input IqN) of the input image received this time is compared with the image quality score (input Iq0) of the input image received last time, and if IqN is higher than Iq0 (IqN> Iq0), the image The evaluation result may be output to the selection unit 40. If IqN is equal to or less than Iq0 (IqN ≦ Iq0), the re-input command 101 may be output to the image input unit 210.

  In the present embodiment, the image processing apparatus may include a threshold value storage unit that stores a threshold value of the image quality score.

  FIG. 16 is a block diagram illustrating an example of a configuration in which the threshold storage unit 70 is provided in the image processing apparatus illustrated in FIG. Here, the image determination unit 250a and the image selection unit 40 receive the input image output from the image input unit 210.

  The threshold storage unit 70 stores a predetermined image quality score threshold (referred to as Th).

  When receiving the input image and the first to third determination images, the image determination unit 250a acquires the threshold value stored in the threshold value storage unit 70. Then, the acquired threshold value is compared with the image quality score (IqN) of the received input image and the image quality scores (Iq1, Iq2, and Iq3 (Iq1 <Iq2 <Iq3)) of the first to third determination images. To do. When all the image quality scores of the received input image and the first to third determination images are lower than the acquired threshold value (IqN <Th and Iq3 <Th), the re-input command 101 is output to the image input unit 210. To do. On the other hand, if there is an image with an image quality score higher than the acquired threshold value (IqN> Th or Iq3> Th), an evaluation result indicating an image with a high image quality score is selected from the input image or the third determination image. To the unit 40.

  As described above, the image determination unit 250a continues to receive the input image and the first to third determination images until an image having an image quality score higher than a predetermined threshold is received. Therefore, an image having an image quality level higher than an arbitrary level can be obtained as an output image.

  Note that the method for comparing the threshold value with the image quality scores of the input image and the first to third determination images is not limited to the method described above. For example, the image determination unit 50 outputs the image re-input command 101 when the image quality score of an arbitrary number of images among the input image and the first to third determination images is lower than a threshold value. Also good. Further, the re-input command 101 may be output when the image quality score of an arbitrary number of images among the input image and the first to third determination images is higher than a threshold value.

  Further, the number of thresholds is not limited to one. For example, using two threshold values (Th1 and Th2 (Th1 <Th2)), an arbitrary number of input images and first to third determination images out of the input image and the first to third determination images. If the image quality score of the image is greater than or equal to Th1 and less than Th2, (or less than T1 and greater than or equal to T2), the re-input command 101 may be output.

  In the present embodiment, the image processing apparatus may include an image number storage unit that stores the maximum number of input images to be received.

  FIG. 17 is a block diagram illustrating an example of a configuration in which the image number storage unit 80 is provided in the image processing apparatus illustrated in FIG. Also in this case, the image determination unit 250b and the image selection unit 40 accept the input image output from the image input unit 210.

  The number-of-images storage unit 80 stores the maximum number of input images (N1) that is the maximum number of input images accepted by the image determination unit 250b.

  The image determination unit 250b stores the received number of input images (N2). When the image determination unit 250b determines to output the re-input command 101, the image determination unit 250b acquires the maximum number of input images (N1) from the image number storage unit 80. When the received number of input images (N2) is smaller than the acquired maximum number of input images (N1) (N1> N2), the re-input command 101 is output to the image input unit 210. On the other hand, if the received number of input images (N2) is equal to or greater than the acquired maximum number of input images (N1) (N1 ≦ N2), the evaluation result is sent to the image selection unit 40 without outputting the re-input command 101. Output.

  As described above, an upper limit of the number of times of accepting the input image can be set, and the image selection unit 40 can output the output image within an arbitrary time.

  The condition for determining whether to output the re-input command 101 or output the evaluation result using the number of input images (N2) and the maximum number of input images (N1) is not limited to the above-described conditions. When it is determined to output the re-input command 101, for example, when the number of input images (N2) is larger than the maximum number of input images (N1) (N1 <N2), the re-input command 101 may be output. When it is determined that the re-input command 101 is output, for example, when the number of input images (N2) is equal to the maximum number of input images (N1) (N1 = N2), the re-input command 101 may be output. . Even when it is determined to output the evaluation result, for example, when the number of input images (N2) is smaller than the maximum number of input images (N1) (N1> N2) (or larger (N1 <N2)), The input command 101 may be output. Even when it is determined to output the evaluation result, for example, when the number of input images (N2) is equal to the maximum number of input images (N1) (N1 = N2), the re-input command 101 may be output. .

  In the present embodiment, the image processing apparatus may include an image storage unit that is a storage unit that stores images.

  FIG. 18 is a block diagram illustrating an example of a configuration in which an image storage unit 90 is provided in the image processing apparatus illustrated in FIG. Also in this case, the image determination unit 250c and the image selection unit 40 accept the input image output from the image input unit 210.

  The image storage unit 90 stores a reference image serving as a reference when evaluating the image quality levels of the input image and the first to third determination images, and the image quality score of the reference image.

  When the image determination unit 250c outputs the re-input command 101 for the first time, the image having the highest image quality score among the received input image and the first to third determination images is set as a reference image, and the reference image and its The image quality score (Iq4) of the reference image is stored in the image storage unit 90. Then, when outputting the re-input command 101 for the second and subsequent times, the image determination unit 250c first acquires the reference image stored in the image storage unit 90 and the image quality score (Iq4) of the reference image. . Then, the image quality score (Iq4) of the acquired reference image is compared with the image quality scores (IqN, Iq1, Iq2, Iq3, (Iq1 <Iq2 <Iq3)) of the input image and the first to third determination images. To do. When the image quality score (IqN) of the input image or the image quality score (Iq3) of the third determination image is higher than the image quality score (Iq4) of the acquired reference image (Iq4 <IqN or Iq4 <Iq3) The image having the highest score and the image quality score (IqN or Iq3) are stored in the image storage unit 90 as the reference image and the image quality score of the reference image. On the other hand, when the image quality score (IqN) of the input image or the image quality score (Iq3) of the third determination image is equal to or less than the image quality score (Iq4) of the acquired reference image (Iq4 ≧ IqN or Iq4 ≧ Iq3) The acquired reference image and the image quality score (Iq4) of the image are stored in the image storage unit 90 as they are.

  When the image determination unit 250c determines to output the evaluation result, the image determination unit 250c acquires the reference image and the image quality (Iq4) of the reference image from the image storage unit 90. When the image quality score (Iq4) of the acquired reference image is higher than the image quality score of the image having the highest image quality score among the input image and the first to third determination images (Iq4> IqN or Iq4> Iq3). ), And outputs the acquired reference image to the image selection unit 40. At this time, the image selection unit 40 sets the reference image output from the image determination unit 50 as an output image. On the other hand, when the image quality score of the acquired reference image is equal to or less than the image quality score of the input image or the first to third determination images (Iq4 ≦ IqN or Iq4 ≦ Iq3), the input image and the first to first images The evaluation result indicating the image having the highest image quality score among the three determination images is output to the image selection unit 40.

  When the image determination unit 250c first stores the reference image in the image storage unit 90, the image storage unit displays an image other than the image with the highest image quality score among the input image and the first to third determination images. 90 may be stored. An arbitrary reference image may be stored in the image storage unit 90 in advance.

  Further, when the image determination unit 250c outputs the re-input command 101 for the second time or later, the input image having the image quality score higher (or lower) than the image quality score of the acquired reference image and the first to third determinations When there are a predetermined number of images, any one of the input images and the first to third determination images may be stored in the image storage unit 90.

  The number of reference images stored in the image storage unit 90 may be any number. For example, among the input image and the first to third determination images, the top two images with the highest image quality scores are stored. May be.

  If the image determination unit 250c determines to output the evaluation result, the image quality score (Iq4) of the reference image acquired from the image storage unit 90 is the highest image quality score among the input image and the first to third determination images. The reference image may be output to the image selection unit 40 when the image quality score is lower than the image quality score of the high image.

  Here, although the processing time increases, as a method for obtaining an output image with a higher image quality level, the reference images stored in the image storage unit 90 are not corresponding to the first to third determination images but to the respective images. The first to third processed images may be used.

  FIG. 19 is a block diagram illustrating another example of a configuration in which an image storage unit 90 is provided in the image processing apparatus illustrated in FIG. As illustrated in FIG. 19, the image determination unit 250d includes not only outputs from the first to third determination image processing units 30-1 to 30-3 but also the first to third image processing units 20-1. Also accepts output from ~ 20-3.

  In the image processing apparatus shown in FIG. 19, for example, when the image determination unit 250d evaluates that the image quality score of the first determination image is the highest, the first processed image corresponding to the first determination image is used as the reference image. Is stored in the image storage unit 90. In this case, the image determination unit 250d may store the image quality score of the first processed image in the image storage unit 90 together with the first processed image instead of the image quality score of the first determination image. .

  Thereby, when the image input unit 210 receives the input of the image a plurality of times, the image determination unit 250d displays the input image, the first to third determination images, and the first to third processed images. An image with the highest image quality score can be selected from among them, and an output image with a higher image quality level can be obtained.

  As an example, the basic configuration of the image processing apparatus in the present embodiment is the configuration of the image processing apparatus shown in FIG. 1 in the first embodiment. However, the image processing apparatus according to the present embodiment can be applied even when the configuration of the image processing apparatuses according to the second and third embodiments described above is used as a basis.

  In the first to fourth embodiments described above, the case where there are three image processing units and determination image processing units has been described as an example, but the number of image processing units or determination image processing units is three. It is not limited.

  In the first to fourth embodiments described above, the first to third image processes are not limited to the smoothing process, the contraction process, and the expansion process described above, and may be image processes of various other methods. Yes, or a combination of some of them.

  In addition, the image quality level evaluation method performed by the image determination unit in the first to fourth embodiments described above is not limited to the template matching described above, and other various evaluation methods may be used, or some of them may be combined. Evaluation method may be used.

  In the present invention, the processing in the image processing apparatus is recorded on a recording medium readable by the image processing apparatus in addition to the above-described dedicated hardware. The program recorded on the recording medium may be read by the image processing apparatus and executed. The recording medium that can be read by the image processing apparatus refers to a transfer medium such as a flexible disk, a magneto-optical disk, a DVD, and a CD, as well as an HDD built in the image processing apparatus. The above program can also be provided through a communication line.

DESCRIPTION OF SYMBOLS 1 Noise with high density value 2 Noise with low density value 10,210 Image input part 11a-11e Input image 20-1-20-3, 20a-1-20a-3 1st-3rd image process part 21a-1 21a-3, 21b-1 to 21b-3, 21c-1 to 21c-3, 21d-1 to 21d-3, 21e-1 to 21e-3 First to third processed images 30-1 to 30 -3 First to third determination image processing units 31a-1 to 31a-3, 31b-1 to 31b-3, 31c-1 to 31c-3, 31d-1 to 31d-3, 31e-1 to 31e -3 First to third determination images 40, 40a Image selection unit 50, 150, 250, 250a, 250b, 250c, 250d Image determination unit 60-1 First preprocessing unit 60-2 Second preprocessing Part 61-1 First preprocessed image 61-2 Second Preprocessed image 70 threshold value storage unit 80 the image number storage unit 90 the image storage unit 101 re-enter instruction 201 executed instruction

Claims (9)

Receiving an input of an image, generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image, and the image quality level of the generated plurality of processed images is An image processing apparatus that selects a processed image that satisfies a predetermined condition ,
An image input unit that receives an input of an image and outputs the received image as an input image;
Receiving an input image output from the image input unit, generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image, and generating the plurality of processed images An image processing unit for outputting an image;
An input image output from the image input unit is received, and the received input includes a plurality of determination image processes associated with each of the plurality of different types of image processing and having a smaller processing amount than the image processing. A determination image processing unit configured to generate a plurality of determination images associated with each of the plurality of processed images and to output the generated plurality of determination images;
The plurality of determination images output from the determination image processing unit are received as evaluation target images whose image quality levels are evaluated, the image quality levels of the received plurality of determination images are evaluated, and the plurality of determinations An image determination unit that outputs an evaluation result indicating a determination image whose image quality level satisfies a predetermined condition among the images for use;
A plurality of processed images output from the image processing unit and an evaluation result output from the image determination unit are received, and a determination image indicated by the received evaluation result among the plurality of received processed images An image selection unit for selecting a processed image corresponding to
A preprocessed image obtained by receiving an input image output from the image input unit and performing a preprocess for clarifying a difference in image quality level between the plurality of determination images with respect to the received input image. A pre-processing unit that generates and outputs the generated pre-processed image,
The determination image processing unit receives an input image output from the image input unit and a preprocessed image output from the preprocessing unit, and determines a predetermined one of the plurality of determination image processes. Performing image processing on the received input image, and performing image processing for determination other than the predetermined one image processing for determination among the plurality of image processing for determination on the received preprocessed image, An image processing apparatus that generates a plurality of determination images. The image processing apparatus according to claim 1 .
The image processing apparatus, wherein the preprocessing unit generates the preprocessed image by randomly adding noise to the received input image. The image processing apparatus according to claim 1 or 2 ,
The image determination unit outputs a re-input command for accepting image input again according to the image quality level of the evaluation target image,
When the image input unit receives a re-input command output from the image determination unit, the image input unit receives the input of the image again and outputs the received image as the input image. The image processing apparatus according to claim 3 .
Having a storage unit,
The image determination unit stores an evaluation target image satisfying a predetermined image quality level among the evaluation target images in the storage unit as a reference image serving as a reference when evaluating the image quality level, and then further When the evaluation target image is received, the reference image stored in the storage unit is acquired, and the evaluation result is obtained by comparing the image quality level of the acquired reference image with the received image quality level of the evaluation target image. Or select and output one of the reference images,
An image processing apparatus that selects the received reference image when the image selection unit receives the reference image output from the image determination unit. The image processing apparatus according to any one of claims 1 to 4 ,
The image determination unit receives an input image output from the image input unit, adds the received input image as the evaluation target image,
The image selection unit receives the input image output from the image input unit, and among the received input image and the plurality of received processed images, the input image corresponding to the image indicated by the evaluation result and the processed image An image processing apparatus for selecting an image. The image processing apparatus according to any one of claims 1 to 5 ,
The plurality of different types of image processing is image processing in which a method for removing noise included in an image is different from each other,
The image processing apparatus, wherein the image quality level is evaluated according to the number of noises included in the image. Processing to accept image input ;
A process of generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image;
A selection process for selecting a processed image whose image quality level satisfies a predetermined condition among the plurality of generated images generated;
An image processing method in an image processing apparatus for execution,
The selection process includes
Each of the plurality of processed images is associated with each of the plurality of different types of image processing, and is subjected to a plurality of determination image processings having a smaller processing amount than the image processing. A determination image generation process for generating a plurality of determination images, which are evaluation target images that are associated and whose image quality level is evaluated;
A process of evaluating image quality levels of the plurality of generated determination images;
A process of selecting the processed image corresponding to a determination image whose image quality level satisfies a predetermined condition among the plurality of determination images;
Including
The determination image generation process includes:
Processing for generating a preprocessed image on which the preprocessing for clarifying the difference in image quality level between the plurality of determination images is performed on the input image;
A predetermined one of the plurality of determination image processes is applied to the input image, and a determination image process other than the predetermined one of the plurality of determination image processes is performed on the input image. Processing to generate the plurality of determination images by applying to a preprocessed image;
Including an image processing method. The image processing method according to claim 7 .
An image processing method further comprising a process of receiving an input of an image again according to an image quality level of the evaluation target image. In the image processing device,
A procedure to accept image input ,
A procedure for generating a plurality of processed images obtained by performing a plurality of different types of image processing on the received input image;
A selection procedure for selecting a processed image whose image quality level satisfies a predetermined condition among the plurality of generated images generated;
An image processing program for causing execution,
The selection procedure is:
Each of the plurality of processed images is associated with each of the plurality of different types of image processing, and is subjected to a plurality of determination image processings having a smaller processing amount than the image processing. A determination image generation procedure for generating a plurality of determination images, which are evaluation target images that are associated and whose image quality level is evaluated;
A procedure for evaluating image quality levels of the plurality of generated determination images;
A procedure for selecting the processed image corresponding to a determination image whose image quality level satisfies a predetermined condition among the plurality of determination images;
Including
The determination image generation procedure includes:
A procedure for generating a preprocessed image that has been subjected to preprocessing for clarifying the difference in image quality level between the plurality of determination images for the input image;
A predetermined one of the plurality of determination image processes is applied to the input image, and a determination image process other than the predetermined one of the plurality of determination image processes is performed on the input image. A procedure for generating the plurality of determination images by applying to a preprocessed image;
Including an image processing program.