Image reading device, image reading method, and image forming apparatus

Abstract

To provide an image reading device that can effectively use information on an image read from a sheet.SOLUTION: An image reading device 1 comprises a reading unit 2, a storage unit 7, a prediction unit 5, and a filling unit 6. The reading unit 2 reads, as an input image P, an image P including an item I consisting of a plurality of elements. The storage unit 7 stores a missing part M in which the elements are partially missing in the item I, and a non-missing part N in which the elements are not missing in the item I. The prediction unit 5 predicts missing elements that are missing in the missing part M on the basis of the form of the item I in the non-missing part N stored in the storage unit 7. The filling unit 6 fills the missing part M with the missing elements predicted by the prediction unit 5.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an image reading apparatus and an image reading method. [Background technology]

[0002] A conventional image reading device is described in Patent Document 1. The image reading device described in Patent Document 1 extracts text information in an area specified by the user from an image read from a sheet, and makes the text information available for later use. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Korean Patent Publication No. 10-2016-0097394 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when extracting text information from an image scanned from a sheet, the image scanning device described in Patent Document 1 may lose part of the text information that overlaps with the graphic information. Furthermore, when the graphic information overlaps with part of the text information, the text information may become an obstacle to the later use of the graphic information. Therefore, the image scanning device described in Patent Document 1 may not be able to effectively utilize the image information, which is an information asset scanned from a sheet.

[0005] The present invention has been made in view of the above-mentioned problems, and has as its object to provide an image reading device, an image reading method, and an image forming device that can effectively utilize information of an image read from a sheet. [Means for solving the problem]

[0006] According to a first aspect of the present invention, an image reading device includes a reading unit, a storage unit, a prediction unit, and a filling unit. The reading unit reads an image including an item composed of a plurality of elements as an input image. The storage unit stores missing portions in which the elements are partially missing from the item and non-missing portions in which the elements are not missing from the item. The prediction unit predicts missing elements missing from the missing portions based on the shape of the item in the non-missing portions stored in the storage unit. The filling unit fills the missing portions with the missing elements predicted by the prediction unit.

[0007] According to a second aspect of the present invention, an image reading method reads an image including an item composed of a plurality of elements as an input image, stores missing portions in which the elements are partially missing from the item and non-missing portions in which the elements are not missing from the item, predicts missing elements that are missing from the missing portions based on the shapes of the elements in the stored non-missing portions, and fills the missing portions with the predicted missing elements.

[0008] According to a third aspect of the present invention, an image forming apparatus includes an image reading device and an image forming unit. The image reading device is described in the first aspect. The image forming unit forms an image on a recording medium. The image reading device forms a new image by combining the items whose missing elements have been filled by the filling unit. The image forming unit is configured to be able to form the new image formed by the image reading device on the recording medium. [Effects of the Invention]

[0009] According to the present invention, the image reading device, the image reading method, and the image forming device can effectively utilize the information of the image read from the sheet. [Brief explanation of the drawings]

[0010] [Figure 1] 1A and 1B are diagrams illustrating an example of an image read from a sheet by an image reading apparatus according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing an example in which an object, which is a collection of items, is extracted from the image. [Figure 3] FIG. 2 is a block diagram of the image reading device. [Figure 4] FIG. 2 is a diagram illustrating an example of the configuration of a prediction unit of the image reading device. [Figure 5] 10 is a flowchart showing a filling process performed by the image reading device. [Figure 6] 10A and 10B are diagrams showing examples of objects filled by the image reading device. DETAILED DESCRIPTION OF THE INVENTION

[0011] An image reading device 11 according to an embodiment of the present invention will be described below with reference to the drawings. In the drawings, identical or corresponding parts are designated by the same reference numerals, and description thereof will not be repeated. In the following description, terms indicating positions or directions, such as "upper," "lower," "horizontal," and "vertical," may also be used. These terms are used for convenience to facilitate understanding of the embodiment, and are not limited to positions or directions when actually implemented.

[0012] [About the composition of Sheet S] An example of an image P (an example of an input image) read from a sheet S by an image reading device 1 according to an embodiment of the present invention will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of an image P read from a sheet S by the image reading device 1.

[0013] Image P shown in FIG. 1 is document information and includes multiple items I. Item I is a collection of elements (e.g., pixels) that form a certain form (shape, pattern, color, etc.) in image P, and each item I is defined as the smallest unit of a component. Item I is divided into text items It and graphic items Is. Text items It are components whose form is text and that make up words, clauses, etc. Each text item It includes one or more characters. Graphic items Is are components whose form is not text.

[0014] Furthermore, for each item I, a group of one or more items I whose shapes are similar to each other at a certain level or higher is defined as an object B. Object B is divided into text objects Bt, which are a group of text items It, and graphic objects Bs, which are a group of graphic items Is.

[0015] In FIG. 1, a group of text items It, each containing a plurality of text items It, is arranged in the center and the top of an image P. The group of text items It in the center is determined to be similar to each other, and a (first) text object Bt1 is formed. Similarly, the group of text items It in the top is determined to be similar to each other, and a second text object Bt2 is formed. Also in FIG. 1, a triangular graphic object Bs1 is formed by one graphic item Is. The first text object Bt1 and the graphic object Bs1 partially overlap at an overlapping portion K.

[0016] In order to effectively utilize the information of image P read from sheet S, image reading device 1 extracts each object B from image P. Here, "extraction" means that object B is stored in a predetermined storage location (such as a storage device) as independent data separate from image P, and made available for later use. The information of image P is used by the user on an object B-by-object basis. Object B may be configured to be manually selected and extracted by the user, or may be configured to be automatically extracted by performing predetermined processing on image P.

[0017] Next, an example of object B extracted from image P will be described with reference to Figures 1 and 2. Figure 2 is a diagram showing an example of object B extracted from image P. Because text object Bt1 shown in Figure 1 has an overlapping portion K, when it is extracted from image P, as shown in the example of Figure 2, information about text item It included in the overlapping portion K may be missing.

[0018] Hereinafter, a portion missing from text object Bt1 will be referred to as missing portion M, and a portion of text object Bt1 where no text item It (element) is missing will be referred to as non-missing portion N. Furthermore, an element missing from missing portion M will be referred to as missing element E. Hereinafter, with respect to any object B, a portion where no element or information is missing will be referred to as missing portion M, and a portion where no element or information is missing will be referred to as non-missing portion N.

[0019] Furthermore, while the information contained in the overlapping portion K of the text object Bt1 remains in the graphic object Bs1 extracted from the image P, information that should be contained in the overlapping portion K of the graphic object Bs1 (such as color information of the portion overlapping with the text item It) may be missing from the graphic object Bs1. Hereinafter, with respect to the graphic object Bs1, the portion from which graphic information is missing will be referred to as a missing portion M, and the portion of the graphic object Bs1 from which information is not missing will be referred to as a non-missing portion N. Furthermore, an element missing from the missing portion M will be referred to as a missing element E.

[0020] Even when an object B extracted from an image P has a missing part M, the present invention corrects the object B by filling in the information of the missing part M to make the object B usable later.

[0021] [Configuration of image reading device 1] The configuration of the image reading device 1 according to the first embodiment of the present invention will be described with reference to Fig. 3. Fig. 3 is a block diagram of the image reading device 1.

[0022] As shown in FIG. 3, the image reading device 1 includes a reading unit 2, a recognition unit 4, a prediction unit 5, a filling unit 6, an input / output unit 3, a memory unit 7, and a control unit 10. The reading unit 2 is typically a scanner, and optically reads an image P from a sheet S and converts the image P into digital data. The sheet S includes an item I made up of multiple elements (e.g., pixels). The recognition unit 4 is a computer program, and identifies missing portions M and non-missing portions N for each object B (described in detail below).

[0023] The storage unit 7 includes a storage device such as a ROM (Read Only Memory) or a RAM (Random Access Memory). The storage device of the storage unit 7 stores data and computer programs. The storage unit 7 stores the missing portion M and the non-missing portion N, which are data of the image P.

[0024] The prediction unit 5 is a computer program that predicts missing elements E that are missing in the missing portion M based on the form of the non-missing portion N stored in the storage unit 7 (details will be described later). The filling unit 6 is a computer program that fills the missing portion M with the missing elements E predicted by the prediction unit 5.

[0025] The input / output unit 3 is a user interface that has an input function for receiving operations to the image reading device 1 as input, and an output function for outputting various information from the image reading device 1. The input / output unit 3 is, for example, a touch panel display that combines an input function and a display function. The input / output unit 3 receives operations to the image reading device 1 from the control unit 10. The control unit 10 displays the results of the operations to the image reading device 1 as output.

[0026] The control unit 10 includes a processor such as a CPU (Central Processing Unit). The memory unit 7 stores data of the image P read by the reading unit 2, information related to operations received from the input / output unit 3, and computer programs constituting the recognition unit 4, prediction unit 5, and filling unit 6. The processor of the control unit 10 executes the computer programs stored in the memory unit 7 to process the data of the image P stored in the memory unit 7. The control unit 10 then causes the input / output unit 3 to display the results of the processing.

[0027] [Configuration of Prediction Unit 5] An example of the configuration of the prediction unit 5 of the image reading device 1 will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the configuration of the prediction unit 5 of the image reading device 1. As shown in Fig. 4, the prediction unit 5 has a pattern prediction unit 54, a text prediction unit 51, a color prediction unit 52, and a shape prediction unit 53.

[0028] (Pattern prediction unit 54) The pattern prediction unit 54 includes, for example, a pattern analysis processing unit 54b. The pattern analysis processing unit 54b performs pattern analysis processing to analyze the pattern of the shape of the missing portion M using information about the shape of the non-missing portion N. The pattern analysis processing unit 54b may be configured to perform processing using a neural network model obtained in advance by performing machine learning.

[0029] A CNN (Convolutional Neural Network) is typically used as the neural network. CNN is a type of neural network that is widely used in fields such as image recognition. CNN is one method for realizing deep learning. In this embodiment, the CNN extracts feature amounts of the non-missing portion N from information about the shape of the non-missing portion N, and extracts local features of the non-missing portion N as a shape pattern. By using the CNN, the pattern analysis processing unit 54b can recognize that the image P contains a non-missing portion N even if the position of the non-missing portion N in the image P changes, if the non-missing portion N is rotated, or if the non-missing portion N is inverted.

[0030] When a neural network model (CNN model) using CNN as the neural network is constructed, machine learning is performed in advance. For example, a set of input / output relationship data is made up of an image P as input and information about the morphological pattern of the non-missing portion N in the image P as output. The CNN model becomes a neural network model that can estimate the morphological pattern of the non-missing portion N from any image P by performing machine learning in advance using multiple sets of input / output relationship data as training data.

[0031] The pattern analysis processing unit 54b uses a CNN model to perform pattern analysis processing on information related to the shape of the non-missing portion N, and predicts the shape pattern of the missing portion M. The pattern prediction unit 54 predicts the missing element E of the missing portion M based on the shape pattern predicted by the pattern analysis processing unit 54b.

[0032] The pattern prediction unit 54 may further be configured to have a difference matching processing unit. The difference matching processing unit performs difference matching processing to compare two pieces of image P data, a missing portion M and a non-missing portion N, and calculate the difference between the missing portion M and the non-missing portion N.

[0033] The pattern prediction unit 54 may further be configured to have an image intersection processing unit. The image intersection processing unit performs image intersection processing to overlap two pieces of image P data, each consisting of a missing portion M and a non-missing portion N, and to detect common portions and relative positions of the images P from the overlapping results.

[0034] The pattern prediction unit 54 may further include an image search unit 55. The image P search unit 55 performs image search processing to search for an image P pattern of the non-missing portion N.

[0035] (Text Prediction Section 51) The text prediction unit 51 includes, for example, a natural language processing unit 51b and a typography analysis processing unit 51c.

[0036] The natural language processing unit 51b performs natural language processing using a natural language processing model (NLP model) obtained in advance through machine learning. An NLP model is a mathematical model capable of performing tasks such as generating, classifying, translating, and summarizing text. An NLP model is obtained in advance through machine learning of the statistical characteristics of text. NLP models include statistical models and neural network models. By using the NLP model, the natural language processing unit 51b predicts the content of a text item It included in a missing portion M of the text object Bt1 from the characteristics of a text item It included in a non-missing portion N of the text object Bt1.

[0037] The typography analysis processing unit 51c analyzes elements including the shape, size, placement, font, color, background, and the like of the text. The typography analysis processing unit 51c analyzes the shape of the text in the non-missing portion N of the text object Bt1 and predicts the shape of the text in the missing portion M of the text object Bt1. The typography analysis processing unit 51c may be configured to perform this in combination with OCR (Optical Character Recognition) processing. By combining OCR processing, such as edge detection and shape recognition, with typography analysis processing, the typography analysis processing unit 51c can analyze the shape of the text more accurately.

[0038] The text prediction unit 51 predicts the missing element E of the text object Bt1 based on the content of the text predicted by the natural language processing unit 51b and the form of the text predicted by the typography analysis processing unit 51c.

[0039] 2 again, the missing portion M of the text object Bt1 specifically includes a text-missing portion Mt where a part of the text item It is missing. Furthermore, the non-missing portion N of the text object Bt1 includes a non-missing text portion Nt where no text item It is missing. The text prediction unit 51 may be configured to predict the missing text portion Mt first, and then predict the missing portion M of the text object Bt1. In other words, the text prediction unit 51 may be configured to predict the missing text portion Mt first, using information about the shape of the non-missing text portion Nt.

[0040] (Color prediction unit 52) The color prediction unit 52 includes, for example, a color matching processing unit 52b. The color matching processing unit 52b performs color matching processing to compare and analyze the colors of the missing portion M and the non-missing portion N of the item I, and predicts the color of the missing portion M. The color prediction unit 52 predicts the missing element E based on the color of the missing portion M predicted by the color matching processing unit 52b.

[0041] The color prediction unit 52 may further be configured to include a smart paint drop unit. The smart paint drop unit accepts a user selection of an element in the non-missing portion N and a user operation of dragging and dropping the selected element into the missing portion M. The color prediction unit 52 predicts a missing element E based on the element selected by the user.

[0042] (Shape prediction unit 53) The shape prediction unit 53 includes, for example, a shape fitting processing unit 53b and a shape matching processing unit 53c. The shape fitting processing unit 53b performs shape fitting processing to predict the shape of the missing portion M of the graphical object Bs1 from the shape of the non-missing portion N of the graphical object Bs1. The shape fitting processing unit 53b may be configured using a CNN model. By using the CNN model, the shape fitting processing unit 53b extracts local shape features of the non-missing portion N of the graphical object Bs1, thereby predicting the shape of the missing portion M of the graphical object Bs1.

[0043] The shape matching processing unit 53c performs a shape matching process that analyzes the similarity between the shapes of the missing portion M of the graphic object Bs1 and the non-missing portion N of the graphic object Bs1. The shape matching processing unit 53c analyzes the similarity of the shapes and extracts shape features such as the contours of the shapes, thereby predicting the shape of the missing portion M of the graphic object Bs1.

[0044] The shape prediction unit 53 predicts the missing element E of the graphic object Bs1 based on the shape of the missing portion M of the graphic object Bs1 predicted by the shape fitting processing unit 53b and the shape matching processing unit 53c.

[0045] [Configuration of the recognition unit 4] Although not shown, the identification unit 4 in Figure 3 may be configured to identify the missing portion M and the non-missing portion N using at least one of the natural language processing unit 51b (natural language processing model) described above and an OCR processing unit (optical character recognition).

[0046] As described above with reference to FIG. 4 , the natural language processing unit 51b of the prediction unit 5 performs natural language processing using an NLP model. The NLP model is capable of generating and classifying text as well as analyzing the meaning and grammar of the text. The continuity of the text item It is determined based on whether each word, sentence, etc. in the text is appropriately connected semantically and grammatically. The natural language processing unit 51b can evaluate the continuity of the text item It by performing natural language processing on the text object Bt1 using the NLP model. Therefore, the natural language processing unit 51b can identify a portion of the text that is semantically or grammatically continuous as a non-missing portion N and a portion that is not continuous as a missing portion M. The natural language processing unit 51b may be shared by the recognition unit 4 and the prediction unit 5.

[0047] The OCR processing unit used by the classification unit 4 performs OCR processing to recognize the text portion of the image P data and convert it into character data. The OCR processing involves image processing to extract character outlines and feature values ​​from the image P, and character recognition to identify characters from the extracted feature values. By extracting the text outlines during the image processing process, the OCR processing unit can extract text-missing portions Mt, where part of the text item It is missing, and text-non-missing portions N, where part of the text item It is not missing.

[0048] As described above, the recognition unit 4 distinguishes between the missing portion M and the non-missing portion N. This saves the user the trouble of manually selecting the missing portion M and the non-missing portion N each time. As a result, the image reading device 1 can effectively and efficiently utilize the information in the image P read from the sheet S.

[0049] [Processing by image reader 1] Next, the processing of the image reading device 1 will be described with reference to Fig. 5. Fig. 5 is a flowchart of the processing of the image reading device 1.

[0050] As shown in FIG. 5, the image reading device 1 first accepts the selection of an object B whose missing element E is to be filled (step S11). When the object B to be filled is selected by the user, the image reading device 1 accepts the selection of whether to automatically or manually identify the missing portion M (step S12). Note that the non-missing portion N is also identified in step S12. The "missing portion M and non-missing portion N" may be abbreviated as "missing portion M." Automatic identification is automatic identification by the identification unit 4. Manual identification is identification performed by manual selection by the user. As described above, the image reading device 1 can automatically identify the missing portion M if the object B to be filled is a text object Bt1. If automatic identification is selected (Yes in step S12), the missing portion M is identified by the identification unit 4 (step S13). Then, the missing portion M is marked (step S14). If manual identification is selected (No in step S12), the missing portion M is identified (selected) by the user (step S14). Then, the missing portion M is marked (step S14). After the missing portion M is marked, the image reading device 1 creates a work area for filling the item I (step S16). Then, the image reading device 1 creates a copy of the object B to be filled in the work area (step S17).

[0051] Steps S18, S19, S20, and S21 are performed on the copied object B, and the prediction unit 5 predicts the missing element E in the missing portion M of object B. Specifically, in step S18, the text prediction unit 51 predicts the missing element E in the missing portion M of the text object Bt1. In step S19, the color prediction unit 52 predicts the missing element E of the color of the missing portion M. In step S20, the shape prediction unit 53 predicts the missing element E in the missing portion M of the graphics object Bs1. In step S21, the pattern prediction unit 54 predicts the missing element E of the pattern in the shape of the missing portion M. Steps S18, S19, S20, and S21 may be performed in parallel.

[0052] The missing element E predicted by the prediction unit 5 is then filled into the item I by the filling unit 6 (step S22). The image reading device 1 accepts a user's selection as to whether or not to create another object B to be filled (step S23). If another object B to be filled is to be created (Yes in step S23), the process returns to step S17. The image reading device 1 then creates a copy of the object B selected in step S11 in the work area and predicts and fills in the missing element E again. If a new object B is not to be created (No in step S23), the image reading device 1 displays all filled objects B created up to that point on the input / output unit 3 (step S24). Then, when a new object is selected by the user, the image reading device replaces the existing object B with the selected new object B (step S25). The image reading device 1 deletes the copy of object B created in the work area (step S26), and the process ends (END).

[0053] An example of object B that has been corrected by filling in missing element E through the above process is shown in Figure 6. As shown in Figure 6, text object Bt1 has been corrected to form corrected text object Bt3, and graphic object Bs1 has been corrected to form corrected graphic object Bs2. There is no missing part M in corrected text object Bt3 or corrected graphic object Bs2.

[0054] With the above-described configuration and processing, the image reading device 1 predicts missing element E that is missing in object B based on the shape of non-missing portion N in object B, and fills in missing portion M. Therefore, even if object B extracted from image P has missing portion M, object B is corrected by filling in missing element E. The user can use the corrected object B. As a result, the image reading device 1 can effectively utilize the information in image P read from sheet S.

[0055] In the image reading device 1, the prediction unit 5 performs pattern analysis, which enables the prediction unit 5 to predict the missing element E with high accuracy. In addition, in the image reading device 1, the prediction unit 5 predicts the missing element using a natural language processing model, which enables the prediction unit 5 to predict the missing element E with high accuracy. In addition, in the image reading device 1, the prediction unit 5 predicts the missing element using color matching processing, which enables the prediction unit 5 to predict the missing element E with high accuracy.

[0056] Therefore, the object B is corrected to its original form with high accuracy. As a result, the image reading device 1 can effectively utilize the information of the image P read from the sheet S.

[0057] [Image forming apparatus 100] Referring again to FIG. 3, as shown in FIG. 3, the image reading device 1 may be configured to be included in an image forming device 100 having an image forming unit 20. The image reading device 1 may be configured to be able to form a new image using an item I in which a missing element E has been filled. The image forming unit 20 forms an image P on a recording medium such as paper. The image forming unit 20 forms the image P read by the image reading device 1, a new image formed by the image reading device 1, etc. on the recording medium.

[0058] An example of the image forming apparatus 100 is a multifunction peripheral such as an MFP (Multi-Function Peripheral). When the image reading apparatus 1 is provided as part of the multifunction peripheral that is the image forming apparatus 100, the image reading apparatus 1 can be used for general purposes, such as printing the read image P, a new image that has been formed, etc. on paper. As a result, the image reading apparatus 1 can effectively utilize the information of the read image P.

[0059] The embodiments of the present invention have been described above with reference to the drawings. However, the present invention is not limited to the above embodiments and can be embodied in various forms without departing from the spirit and scope of the present invention. The drawings mainly show each component in a schematic manner for ease of understanding, and the thickness, length, number, spacing, etc. of each component shown in the drawings may differ from the actual ones due to the convenience of creating the drawings. Furthermore, the materials, shapes, dimensions, etc. of each component shown in the above embodiments are merely examples and are not particularly limited, and various modifications are possible within a scope that does not substantially deviate from the configuration of the present invention. [Industrial Applicability]

[0060] The present invention is useful in the fields of, for example, image reading devices, image reading methods, and image forming devices, and has industrial applicability. [Explanation of symbols]

[0061] S seat P Image I Item It text item Is Shape Item Bt Text Object Bs Shape Object M Missing part N non-missing part K overlapping part 1. Image reader 2 Reading unit 3 Input / output section 4 Identification unit 5. Prediction Department 6 Filling section 7 Memory section 10 Control Unit 51 Text Prediction 52 Color Prediction Unit 53 Shape Prediction Unit 54 Pattern Prediction Unit 51b Natural Language Processing Unit 51c Typography Analysis Processing Unit 52b Color matching processing section 53b Shape fitting processing section 53c Shape matching processing section 54b Pattern analysis processing unit

Claims

1. a reading unit that reads an image including an item composed of a plurality of elements as an input image; a storage unit that stores a missing portion in which the element is partially missing from the item and a non-missing portion in which the element is not missing from the item; a prediction unit that predicts missing elements that are missing in the missing portion based on the form of the items in the non-missing portion stored in the storage unit; a filling unit that fills the missing portion with the missing element predicted by the prediction unit; An image reading device comprising:

2. The image reading device according to claim 1 , further comprising an identification unit that identifies the missing portion and the non-missing portion.

3. the prediction unit predicts the missing elements using a neural network model obtained in advance by performing machine learning; 3. The image reading device according to claim 1, wherein the neural network model performs a pattern analysis process for analyzing a pattern of the shape of the missing portion using information about the shape of the non-missing portion.

4. the input image, the item including text; The image reading device according to claim 2 , wherein the identification unit identifies the missing portion and the non-missing portion using at least one of a natural language processing model obtained in advance by machine learning and optical character recognition.

5. the input image, the item including text; 5. The image reading device according to claim 1, wherein the prediction unit predicts the missing elements using a natural language processing model obtained in advance by machine learning.

6. The image reading device according to claim 1 or 2, wherein the prediction unit predicts the color of the missing element using a color matching process that compares and analyzes the colors of the missing portion and the non-missing portion of the item.

7. An image containing an item composed of multiple elements is read as an input image, storing a missing portion in which the element is partially missing from the item and a non-missing portion in which the element is not missing from the item; predicting missing elements that are missing in the missing portion based on the stored morphology of the elements of the non-missing portion; and filling the missing portion with the predicted missing element.

8. The image reading device according to claim 1 or 2; an image forming unit that forms an image on a recording medium; Equipped with the image reading device combines the items whose missing elements have been filled by the filling unit to form a new image; The image forming unit is an image forming apparatus capable of forming the new image formed by the image reading device on the recording medium.

Images