JP2006059353A - Electronic album device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow a desired image to be easily retrieved regardless of the number of thumbnails in an electronic album device capable of selecting the desired image from a plurality of images to display it on a screen. <P>SOLUTION: An electronic album device 1 calculates feature quantities of thumbnails of a plurality of original images stored in a CD-RW 5 in response to an input instructing classification of images and classifies similar original images into common groups on the basis of calculation results and outputs a plurality of original images belonging to one group to a monitor 2 in association with each other. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an electronic album apparatus capable of storing a plurality of images taken by a digital camera or the like and selecting a desired image from these images and displaying it on a screen. is there.

In a digital still camera, image data of still images obtained by shooting can be recorded on a memory card. However, the number of still images that can be recorded on one memory card is small, and a large number of still images can be recorded. In order to save, a plurality of memory cards are required.
Therefore, by writing a plurality of still images recorded on a memory card into a data storage device such as a CD-R (Compact Disc-Recordable) or CD-RW (Compact Disc-Rewritable), a large number of still images can be recorded. An electronic album apparatus has been developed that can be stored and can be searched for a large number of still images.

  In this type of electronic album device, a plurality of still images (original images) are stored together with respective thumbnails (reduced images) in the data storage device. For example, a plurality of still images stored in the data storage device are stored. When a desired still image is displayed from an image on an image display device such as a liquid crystal display, a selection screen including thumbnails of each image is displayed first, and then the user selects a target image from the selection screen. A method of displaying a target image large on the screen by selecting a thumbnail of the image is employed. According to this method, since the selection screen can be displayed quickly, the user can display a desired image in a short time.

However, in the conventional electronic album apparatus, the user has to perform the operation of selecting a target image thumbnail from a plurality of thumbnails, and this operation becomes difficult as the number of thumbnails increases.
Accordingly, an object of the present invention is to provide an electronic album device that can easily search for a desired image regardless of the number of thumbnails.

  An electronic album apparatus according to the present invention includes an input unit for instructing classification of a plurality of images stored in a data storage device, and thumbnails of the images stored in the data storage device in response to the input of the command Based on the calculation means for calculating the feature amount of the thumbnail, and a plurality of images stored in the data storage device as targets, a similar group of images similar to each other based on the feature amount of each thumbnail And classifying means for associating a plurality of images belonging to the same group with each other and outputting them to the image display device.

In the electronic album device of the present invention, when searching for a desired image from a plurality of images stored in the data storage device, a command for classifying the plurality of images is input. In response to the input of the command, the electronic album device calculates the geometric feature amount of the thumbnail of each image for a plurality of images stored in the data storage device, and then collates the feature amounts with each other. Thus, the plurality of images stored in the data storage device are classified into a plurality of groups.
As a result, the plurality of images stored in the data storage device are grouped by a plurality of images having a high degree of similarity, and are displayed on the screen of the image display device for each group.

  Accordingly, the user can first significantly reduce the number of images to be selected by selecting a group to which the target image should belong, and then select the target image from a plurality of images belonging to the group. Therefore, it is possible to reduce the labor of image search as compared with the conventional case.

  According to the electronic album device of the present invention, a desired image can be easily searched regardless of the number of thumbnails.

Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings.
As shown in FIG. 1, a monitor (2) is connected to the electronic album apparatus (1) according to the present invention, and an image can be displayed. The user can instruct the electronic album apparatus (1) to perform various operations by operating a remote controller (hereinafter abbreviated as a remote controller) (3).
Further, the electronic album device (1) is provided with a memory card interface (13) and a CD-RW interface (14), and the memory card interface (13) is photographed using a digital camera (not shown). Is loaded with a memory card (4) in which a plurality of images (still images and / or moving images) are recorded, and the CD-RW interface (14) can store a large number of images. Is loaded.

  The electronic album apparatus (1) includes a CPU (6), a ROM (7) in which computer programs for realizing various operations are registered, a RAM (8) as a working memory, and a monitor ( 2) a display control circuit (9) for controlling the operation of the memory card, a memory card control circuit (10) for controlling the operation of the memory card interface (13), and a CD-RW for controlling the operation of the CD-RW interface (14). A control circuit (12) and a receiver (11) for receiving a command signal (infrared signal) from the remote controller (3) are provided.

  By loading the memory card (4) taken out from the digital camera into the memory card interface (13) of the electronic album device (1), a plurality of original images and thumbnails written on the memory card (4) are read out. And stored in the CD-RW (5) loaded in the CD-RW interface (14).

  For example, as shown in FIG. 4, in the CD-RW (5), a large number of original images are stored together with their respective thumbnails. When a desired image is searched from these original images, the purpose of the search is as follows. Enter the thumbnail of the image (search image) to be used. Since the thumbnail is created by photographing with a digital camera and stored in the memory card (4), the thumbnail of the search image is read from the memory card.

  In the search mode, the electronic album device (1) first calculates the feature amount of the thumbnail of the search image, and then, for the plurality of original images stored in the CD-RW (5), the feature amount of the thumbnail of each original image. Are sequentially calculated, and the calculation result is compared with the feature amount of the thumbnail of the search image to find an image having the highest similarity to the search image.

Here, for the calculation and collation of the feature amount of the thumbnail, for example, nonlinear normalization (“Linear density equalization-nonlinear normalization method for correlation method” IEICE Transactions '84 / 11 Vol.J67-D No. 11), then use the method of matching the direction pattern (“Analysis of Handwritten Kanji by Direction Pattern Matching Method” IEICE Transactions '82 / 5 Vol.J65-D No.5) I can do it. That is, binarization is performed based on the density of the image, nonlinear normalization is performed, an edge pattern of the image is obtained, and a feature amount obtained by mapping the edge direction component and position in a feature space for each of the four directions is obtained. The matching degree (similarity) can be calculated by performing pattern matching of two feature quantities to be matched ("Analysis by hand pattern kanji direction pattern matching method" IEICE Transactions '82. / 5 Vol.J65-D No.5).
The image searched in this way is supplied to the monitor (2) in FIG. 1 and displayed on the screen.

  The search image is not limited to an image obtained by photographing a real object with a digital camera, but may be an image obtained by photographing a drawing depicting an image of the search image as shown in FIG.

Further, the image stored in the CD-RW (5) may be not only a still image but also a moving image. For example, as shown in FIG. Is composed of 10 image planes 1 to 10 and image 13 is composed of 20 image planes 1 to 20, the thumbnail of the moving image is one out of 10 image planes constituting the moving image. The thumbnails of the image planes extracted at the rate of. Similarly in this case, the thumbnail of the search image is collated with the thumbnail of each image (still image and moving image) by the feature amount.
In order to determine a thumbnail of a moving image, for example, a method of detecting an image plane whose feature amount has changed beyond a threshold in a series of image planes constituting the moving image and selecting the image plane as a thumbnail can be employed.

FIG. 2 shows an image search procedure based on the search image.
First, in step S1, the thumbnail S ′ of the search image S is read, and in step S2, the feature quantity FS of the read thumbnail S ′ is calculated. Next, in step S3, after setting the image number N to 0, in step S4, the thumbnail N ′ of the image N is read, and in step S5, the feature amount FN of the thumbnail N ′ of the image N is calculated. To do.

Thereafter, in step S6, a matching degree (similarity) RN between the feature quantity FN of the thumbnail of the image N and the feature quantity FS of the thumbnail S ′ of the search image S is calculated.
Subsequently, in step S7, it is determined whether or not N is 0. If it is determined as YES, the process proceeds to step S8, where M = N and RM = RN. If NO, the process proceeds to step S9 to determine whether or not RN exceeds RM. If YES is determined in this step, the process proceeds to step S10 and M = N and RM = RN are set. . If no, step S10 is bypassed.
Thereafter, in step S11, the image number N is counted up, and in step S12, it is determined whether N has exceeded the maximum image number. If it is determined NO, the process returns to step S4. In the case of yes, it transfers to step S13, outputs the image M as an image (search result) with the highest collation degree, and complete | finishes a procedure.

As a result, an image most similar to the search image is displayed on the monitor (2) shown in FIG.
Therefore, the user can automatically obtain a desired image, and there is no need to select a thumbnail of a target image from a large number of thumbnails as in the past.

Further, the electronic album device (1) according to the present invention calculates the feature amount of each thumbnail for a large number of images stored in the CD-RW (5) in the classification mode, and the feature of the thumbnail of each image. The quantities are collated with each other to classify similar images as a common group, and the result is output.
Here, the above-described nonlinear normalization and direction pattern matching methods can be employed for calculating and collating thumbnail feature values.

  As a result of the classification, for example, as shown in FIG. 7, a large number of original images and thumbnails are grouped and displayed for each group.

FIG. 3 shows the image classification procedure.
First, in step S21, the group number GC is set to 0, and in step S22, the image number N is set to 0. Next, in step S23, it is determined whether or not N exceeds the maximum image number. If NO is determined here, the process proceeds to step S24, where the thumbnail N 'of the image N is read, and step S25 is performed. Then, the feature amount FN of the thumbnail N ′ of the image N is calculated.

  Thereafter, in step S26, it is determined whether or not N is 0. If YES is determined in this step, the process proceeds to step S37, in which the group element number GEN [GC] [0] is set to N, In step S38, the number of group elements GEN [GC] is set to 1, in step S39, the feature value GF [GC] of the thumbnail of the group representative (first) image is set to FN, and in step S40, the group feature GF [GC] is set. After setting the number GC to 1, N is counted up in step S41, and the process returns to step S23.

If NO is determined in step S26, the process proceeds to step S27, the group number GN is set to 0, and the group of the feature amount FN of the thumbnail N ′ of the image N and the group number GN is set in step S28. The matching degree RG with the feature quantity GF [GN] of the thumbnail of the representative (first) image is calculated.
Next, in step S29, it is determined whether or not the matching level RG is equal to or higher than the threshold value RT of the similar image matching level. If it is determined NO, the process proceeds to step S30 and GN is counted. In step S31, it is determined whether GN is lower than GC. If it is determined YES, the process returns to step S28.

If it is determined as YES in step S29, the process proceeds to step S42, the group element number (addition) GEN [GN] [GEC [GN]] is set to N, and the number of group elements (addition) is set in step S43. After setting GEC [GN] to GEC [GN] +1, N is counted up in step S44, and the process returns to step S23.
If NO is determined in step S31, the process proceeds to step S32 where the group element number GEN [GC] [0] is set to N. In step S33, the group element number GEN [GC] is set to 1. In step S34, the feature amount GF [GC] of the thumbnail of the group representative (first) image is set to FN. Further, in step S35, the group number GC is counted up, and in step S36, N Is counted up and the process returns to step S23.
Thereafter, when it is determined YES in step S23, the procedure is terminated.

By executing the above procedure, the following classification results are obtained.
(1) Number of groups = GC
(2) Number of elements in each group = GEC [GN]
(GN: group number, 0 ≦ GN <GC)
(3) Image number of each group element = GEN [GN] [i]
(i: element number, 0 ≦ i <GEC [GN])

In this way, a large number of images stored in the CD-RW (5) are grouped with high similarity, and images belonging to each group are displayed on the screen of the monitor (2). .
Therefore, a large number of images can be easily organized.
Further, when searching for a target image, first, by selecting a group to which the target image should belong, the number of images to be selected can be greatly reduced, and then from a plurality of images belonging to the group. What is necessary is just to select the target image.

  In addition, each part structure of this invention is not restricted to the said embodiment, A various deformation | transformation is possible within the technical scope as described in a claim. For example, the calculation of the feature quantity of thumbnails, the collation, and the calculation of the similarity are not limited to the method of the above-described embodiment, and various known methods can be employed.

It is a block diagram showing the structure of the electronic album apparatus based on this invention. It is a flowchart showing the procedure of a search process. It is a flowchart showing the procedure of a classification process. It is a figure explaining a search process. It is a figure showing the search image and thumbnail of a handwritten drawing. It is a figure explaining the search process of the image containing a moving image. It is a figure showing the result of a classification | category process.

Explanation of symbols

(1) Electronic album device
(2) Monitor
(3) Remote control
(4) Memory card
(5) CD-RW
(13) Memory card interface
(14) CD-RW interface

Claims (1)

A data storage device in which a plurality of images are stored together with respective reduced images and an image display device are connected, and a desired one or a plurality of images are selected from the plurality of images and reduced images stored in the data storage device. In an electronic album apparatus capable of selecting and displaying an image and / or a reduced image on an image display apparatus,
Input means for commanding classification of a plurality of images stored in the data storage device;
In accordance with the input of the command, calculation means for calculating the feature amount of the reduced image based on the reduced image of each image stored in the data storage device;
For a plurality of images stored in the data storage device, based on the feature amounts of the respective reduced images, classification is performed such that similar images are grouped together, and a plurality of images belonging to the same group are mutually connected. An electronic album device comprising classification means for associating and outputting to an image display device.

Images