KR102447747B1 - Image processing apparatus and method - Google Patents

Abstract

An image processing device and a method thereof, which are capable of quickly merging polygons existing in an image, are suggested. The suggested device comprises: a storage unit which stores, in advance, an area map including information on a polygon area to which each pixel of the image belongs; and a processor which loads instructions stored in the storage unit. Accordingly, the device creates a neighboring relationship matrix including a neighboring relationship coefficient representing neighboring relationship relationships between a plurality of polygon areas included in the image, compares a first polygon area to which a first pixel among the pixels of the image belongs and a second polygon area to which a second pixel adjacent to the first pixel belongs to each other, calculates a neighboring coefficient between the first polygon area and the second polygon area according to the comparison result, creates a neighboring relationship matrix using the calculated neighboring coefficient, merges the polygon areas based on the generated neighboring relationship matrix, and outputs an area map in which the polygon areas are merged.

Description

Image processing apparatus and method

The present invention relates to an image processing apparatus and method, and more particularly, to an apparatus and method capable of detecting polygons in an image and merging the detected polygons.

Coloring an image means assigning a color value to each pixel in the image. On the other hand, if there are shapes in the image, the corresponding shapes may be colored with the same color. To this end, polygons corresponding to each shape are detected, and each of the detected polygons is colored.

On the other hand, as a conventional method for detecting such a polygon, there is a method using a ball 5 of a predetermined size called a trap ball (trapped ball) as shown in FIG. Specifically, the ball 5 is rolled on the image and the ball 5 is detected as one polygon for each rolled area unit. For example, in the case of area 1, when the ball 5 is rolled, it is divided into an area 2 in which the ball 5 can roll and an area 3 in which the ball 5 cannot roll as shown in the right figure of FIG. 1 . Accordingly, the region 2 and the region 3 are preferably detected as one polygon in the same region, but are detected as separate polygons due to the size limit of the ball 5 .

Since color processing is performed for each polygon, a process of merging a plurality of polygons corresponding to the same area as above is required for efficient coloring. That is, in the case of FIG. 1 , it is efficient to merge the polygons 2 and 3 into one polygon to proceed with coloring.

In the conventional case, after selecting a polygon to be merged (eg, the second polygon 2 of region 2) and extracting the boundary point of the selected polygon to be merged, another polygon including the boundary point (i.e., boundary pixel) of the polygon to be merged is selected. detect In the case of FIG. 2 , the third polygon 3 in the region 3 will be detected as another polygon including the boundary point of the polygon 2 to be merged. Merges other detected adjacent polygons to the merge target polygon.

However, since the conventional polygon merging algorithm detects and merges adjacent polygons for each polygon, as the image becomes more complex, the number of polygons increases, which increases the time required to operate the polygon merging algorithm.

The matters described in the background art above are intended to help the understanding of the background of the invention, and may include matters that are not disclosed 　 prior art.

Prior art 1: Republic of Korea Patent Registration No. 10-2316960 (2021.10.19) Prior Art 2: Republic of Korea Patent Registration No. 10-0210400 (April 26, 1999)

The present invention has been proposed in view of the above-described conventional circumstances, and an object of the present invention is to provide an image processing apparatus and method capable of quickly merging polygons existing in an image.

In order to achieve the above object, an image processing apparatus according to a preferred embodiment of the present invention includes: a storage unit which stores in advance a region map including information on a polygon region to which each pixel of an image belongs; and a processor having an arithmetic processing function, wherein the processor loads the instructions stored in the storage unit, thereby generating an adjacency matrix including an adjacency coefficient representing the adjacency relationship between a plurality of polygon regions included in the image. and comparing the first polygon area to which a first pixel of the pixels of the image belongs and a second polygon area to which a second pixel adjacent to the first pixel belongs, and according to a result of the comparison, the first polygon area and the The adjacency coefficient between the second polygon regions is calculated, the adjacency matrix is generated using the calculated adjacency coefficient, and as the generation of the adjacency matrix is completed, polygon regions are divided based on the generated adjacency matrix. Outputs an area map in which the polygon area is merged by merging.

The processor compares the first polygon area and the second polygon area with each other, and determines whether the first polygon area and the second polygon area are different from each other according to a result of the comparison, and the first polygon area and the second polygon area If the second polygon regions are different from each other, a count value may be generated in response thereto, and the adjacency coefficient may be calculated based on the generated count value.

When the comparison of the first polygon area to which the first pixel belongs and the second polygon area to which the second pixel adjacent to the first pixel belongs is finished, the processor is configured to select a pixel adjacent to the right of the first pixel as the first pixel , and a pixel adjacent to the right of the updated first pixel and a pixel adjacent to the bottom of the updated first pixel are set as the second pixel to repeat the comparison.

The processor may calculate the count value based on the number of the second pixels belonging to the second polygon area determined to be different from the first polygon area.

In a preferred embodiment of the present invention, the storage unit may further store a merging condition for merging between two adjacent polygonal regions, and the processor may merge the polygonal regions using the merging condition.

The merging condition may include information on a threshold value serving as a reference value for merging between the two adjacent polygon regions.

The processor may compare the adjacent coefficients corresponding to the two different adjacent polygonal areas with the threshold, and if the adjacent coefficients exceed the threshold according to the comparison result, the two adjacent different polygonal areas may be merged.

The merging condition may further include a size of the polygon area, and the processor may merge the two adjacent polygon areas into a larger polygon area based on the size of the polygon area when merging the two adjacent polygon areas.

The processor generates a merge map indicating a polygon area to be merged based on the generated adjacency matrix as the generation of the adjacency matrix is completed, and generates the generated merge map as the generation of the merge map is completed. By applying to the area map, an area map in which two adjacent polygon areas are merged may be generated.

A preferred embodiment of the present invention further comprises a communication unit for receiving the area map, wherein the communication unit sends an area map transmission completion signal to the processor after sending the received area map to the processor, and the processor may store the area map in the storage unit after sending the area map reception completion signal according to the reception completion of the area map as an ACK signal to the communication unit.

Meanwhile, the image processing method according to a preferred embodiment of the present invention is a method performed in an image processing apparatus including a storage unit and a processor, wherein the storage unit includes an area including information on a polygon area to which each pixel of an image belongs. pre-saving the map; wherein the processor generates an adjacency matrix including an adjacency coefficient representing an adjacency relationship between a plurality of polygon regions included in the image, wherein a first polygon region to which a first pixel of the pixels of the image belongs and the first polygon region A second polygon area to which a second pixel adjacent to one pixel belongs is compared with each other, and an adjacency coefficient between the first polygon area and the second polygon area is calculated according to a result of the comparison, and using the calculated adjacency coefficient, the generating an adjacency matrix; and outputting, by the processor, a polygonal area merged area map by merging polygon areas based on the generated adjacency matrix as the generation of the adjacency matrix is completed.

According to the present invention having such a configuration, it is possible to create and update an adjacency matrix for two different polygon regions adjacent to a region map with one raster scan, and merge based on the nonzero component of the adjacency matrix. After creating the map, you can merge two different polygonal areas adjacent to each other by applying the merge map to the area map.

For this reason, polygon merging can be performed much faster than the existing polygon merging method.

1 is a diagram for explaining general image editing software.
2 is a diagram employed in the conventional polygon merging description.
3 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
FIG. 4 is an internal configuration diagram of the control unit shown in FIG. 3 .
5 and 6 are examples of area maps.
FIG. 7 is a diagram illustrating an example of an adjacency matrix generated by the adjacency matrix generator shown in FIG. 4 .
8 is a flowchart illustrating an image processing method according to an embodiment of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention.

The terms used in the present application are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the possibility of the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. In describing the present invention, in order to facilitate the overall understanding, the same reference numerals are used for the same components in the drawings, and duplicate descriptions of the same components are omitted.

3 is a block diagram of an image processing apparatus according to an embodiment of the present invention.

The image processing apparatus 100 according to an embodiment of the present invention may include a communication unit 10 , a storage unit 20 , and a control unit 30 . If necessary, the communication unit 10, the storage unit 20, and the control unit 30 may also be referred to as a communication module, a storage module, and a control module.

The communication unit 10 may receive a region map from an external device such as a user terminal (not shown). Here, the area map received by the communication unit 10 includes a value of each pixel of the image. The value of each pixel of the image may mean a region id including the corresponding pixel. In this case, the area ID may be referred to as a polygon area ID. The area map received by the communication unit 10 may be referred to as an area map before polygon merging.

That is, the communication unit 10 may receive the area map from the external device in response to the area map transmission signal from the external device, and output the area map reception completion signal according to the completion of the reception of the area map. More specifically, when receiving the area map transmission signal from the external device, the communication unit 10 transmits an area map reception enable signal as an ACK signal to the external device in response thereto. Accordingly, in response to the ACK signal, the external device transmits a predetermined area map to the communication unit 10 and then transmits an area map transmission completion signal. For this reason, the communication unit 10 may transmit the area map reception completion signal according to the reception completion of the predetermined area map as an acknowledgment (ACK) signal to the external device.

Then, the communication unit 10 sends the received area map to the control unit 30 so that the control unit 30 can store the area map in the storage unit 20 . In this case, the communication unit 10 will send the area map transmission completion signal to the control unit 30 after sending the received area map to the control unit 30 . Then, the control unit 30 can store the area map in the storage unit 20 after sending the area map reception completion signal according to the reception completion of the area map to the communication unit 10 as an ACK signal.

If necessary, the communication unit 10 may receive a predetermined sketch image instead of the area map.

The user terminal described above is a computing device having an arithmetic processing function, and may be a mobile device or a fixed device. For example, the external device may mean a computer, a personal computer (PC), a smart phone, a navigation system, a laptop computer, a tablet computer, a wearable device, or a tablet, but embodiments of the present invention are not limited thereto.

The communication unit 10 may communicate with an external device according to a wireless communication network or a wired communication network, but may send a polygon-merged area map to the external device.

For example, a wireless communication type network includes a wireless LAN (WLAN), a wireless broadband (Wibro), a wideband CDMA (WCDMA), an IEEE 802.16, a long term evolution (LTE), LTE-A (Long Term Evolution-Advanced), Bluetooth (Bluetooth), RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee, Near Field Communication (Near) Field Communication (NFC)), ultrasonic communication (Ultra Sound Communication (USC)), visible light communication (VLC), Wi-Fi (Wi-Fi), and the like may be included, but are not limited thereto.

On the other hand, wired communication type networks include wired LAN (Local Area Network), wired WAN (Wide Area Network), Power Line Communication (PLC), USB communication, Ethernet, serial communication, It may include, but is not limited to, optical/coaxial cable communication.

The storage unit 20 may store data necessary for the operation of the device 100 . According to embodiments, the storage unit 20 may store the area map (ie, the area map before the polygon merging) received from the communication unit 10 , and the polygon merged generated by the processing operation in the device 100 . You can save area maps.

Also, the storage unit 20 may store in advance a merging condition for merging between two adjacent polygonal regions. Here, the merging condition may include information on a threshold value that is a reference value for merging between two adjacent polygonal regions. In addition, the merging condition may further include a size of a polygon region that enables merging into a larger polygon region when two adjacent polygon regions are merged.

Also, the storage unit 20 may store a program composed of instructions for performing a series of operations performed by the device 100 .

The storage unit 20 may include a non-volatile memory device or a volatile memory device. According to embodiments, the storage unit 20 may be configured to be included in the control unit 30 , but is not limited thereto.

The controller 30 may control overall operations of the device 100 .

For example, the control unit 30 may store the area map received through the communication unit 10 in the storage unit 20 . That is, the control unit 30 may store the area map in the storage unit 20 in response to the area map reception completion signal from the communication unit 10 .

If the communication unit 10 receives a predetermined image (eg, a sketch image) instead of an area map, the control unit 30 generates (extracts) a polygon based on the image and generates (extracts) a polygon based on the area map (eg, a sketch image). , an area map before merging polygons).

Also, the controller 30 may read the stored area map and perform signal processing for polygon merging, thereby generating and outputting the polygon merged area map.

The controller 30 may refer to a device capable of performing a series of operations or determinations for the operation of the device 100 . For example, the controller 30 may be a central processing unit (CPU), a micro controller unit (MCU), a graphic processing unit (GPU), an application specific integrated circuit (ASIC), or a field programmable gate array (FPGA), The present invention is not limited thereto.

In other words, the control unit 30 includes at least one processor having an arithmetic processing function. The controller 30 may control operations performed in the device 100, and is included in the device 100 so that a predetermined operation (eg, an image processing operation in the present invention; may be referred to as a polygon merging operation) is performed. Other configurations can be controlled. Specifically, the processor included in the control unit 30 executes at least one instruction stored in the storage unit 20 that may be included internally, so that a predetermined operation (eg, image processing operation in the present invention) is performed. can be controlled

In addition, in an embodiment of the present invention, at least one processor included in the controller 30 may control operations performed in the device 100 and may control other components included in the device 100 to perform a predetermined operation. can Accordingly, even if a case in which the controller 30 controls a predetermined operation (eg, an image processing operation in the present invention) to be performed is described as an example, at least one processor included in the control unit 30 performs a predetermined operation (eg, this It will be apparent that the image processing operation in the present invention can be controlled to be performed.

4 is an internal configuration diagram of the control unit shown in FIG. 3 , FIGS. 5 and 6 are examples of area maps, and FIG. 7 is an example of an adjacency matrix generated by the adjacency matrix generator shown in FIG. 4 . the drawing shown.

The control unit 30 includes a polygon generation unit 31 , a region map generation unit 32 , an adjacency matrix generation unit 33 , a merging condition setting unit 34 , a merging map generation unit 35 , and a merging unit 36 . ) may be included. If necessary, the polygon generating unit 31, the area map generating unit 32, the adjacency matrix generating unit 33, the merging condition setting unit 34, the merging map generating unit 35 and the merging unit 36 are The polygon generating module, the area map generating module, the adjacency matrix generating module, the merging condition setting module, the merging map generating module, and the merging module may also be referred to as a merging module.

When a predetermined image (eg, a sketch image) is received from the communication unit 100 , the polygon generator 31 may generate (extract) a polygon based on the image. Here, the polygon may be generated by classifying each of the regions in the image. Each of the generated polygons may include one or more pixels. In this case, the one or more pixels may have a corresponding area ID value.

The polygon generation unit 31 transmits the polygon to the area map generation unit 32 together with the polygon transmission signal as the polygon generation is finished.

As the area map generator 32 receives a polygon, it generates a corresponding polygon reception completion signal and sends it to the polygon generator 31 .

The area map generator 32 may generate an area map (eg, an area map before merging polygons) based on the received polygon.

For example, the area map received by the communication unit 10 or generated by the area map generation unit 32 may have the same shape as in FIG. 5 . The area map may include a corresponding area ID (which may also be referred to as a polygon area ID) for each pixel. In FIG. 5 , polygons of each area are displayed in addition to the area ID for each pixel. This is for the sake of understanding, and in reality, polygons of each area will not be displayed.

If the communication unit 10 receives the area map, the polygon generation unit 31 and the area map generation unit 32 will not be needed.

The adjacency matrix generator 33 may generate an adjacency matrix for two different adjacent areas (ie, a polygon area) while performing a raster scan of the area map as a whole, and may update the generated adjacency matrix. Here, the two adjacent different regions are two regions having different values assigned to the pixels of the corresponding region, respectively, and mean two adjacent regions.

The adjacency matrix may indicate how many boundary pixels are connected to each other for two different adjacent regions (ie, polygon regions).

For example, during raster scan, two different polygon regions adjacent to each other may be detected in multiple numbers, for example, (r_1, r_2), (r_2, r_3), (r_1, r_4), (r_2, r_4). . At this time, the adjacency matrix M for two different adjacent polygon regions r_1 and r_2 may be expressed as M[r_1, r_2], and the other two polygon regions other than the polygon regions r_1 and r_2 are also It can be expressed in this way. In addition, when initially generating the adjacency matrix for two different adjacent regions (eg, r_1, r_2; polygon region), it is recommended to initialize M[r_1, r_2] (that is, set to “0 (zero)”). need. Of course, if there are two different adjacent regions (e.g., r_2, r_3; polygon region), the initialization of M[r_2, r_3] ( That is, it is necessary to set it to "0 (zero)".

In this way, together with the generation of the adjacency matrix, the adjacency matrix generating unit 33 determines whether each of the pixels in the area map is adjacent to another area (ie, polygon area) in the area to which the pixel belongs during the above-described raster scan. is detected, and the detected result may be reflected in the corresponding adjacency matrix. Here, reflection means updating the value of the corresponding adjacency matrix with the detected result.

In other words, if it is assumed that the region map as shown in FIG. 6 is received, the adjacency matrix generator 33 may finally generate the adjacency matrix M in the form of a 4*4 matrix through raster scan. Since the values of the pixels in FIG. 6 are marked up to 1, 2, 3, and 4, it can be seen that a total of 4 areas (ie, polygons) are included. Accordingly, the adjacency matrix generator 33 may generate the adjacency matrix M in the form of a 4*4 matrix.

Then, the adjacency matrix generator 33 scans the entire area map in raster scan order, and detects whether each pixel in the area map is adjacent to another area in the area to which the pixel belongs. Then, an adjacency coefficient corresponding to the detection result may be calculated, and the adjacency relationship matrix M may be updated using the calculated adjacency coefficient (refer to FIG. 7 ).

In other words, the adjacency matrix generating unit 33 generates an adjacency matrix including adjacency coefficients representing adjacencies between a plurality of polygon regions included in the image, The first polygon area and the second polygon area to which the second pixel adjacent to the first pixel belongs are compared with each other, and an adjacency coefficient between the first polygon area and the second polygon area is calculated according to the comparison result, and the calculated adjacency coefficient is used. Thus, an adjacency matrix can be created. In this case, the first pixel (that is, the pixel currently selected through raster scan) and the second pixel (that is, an unselected pixel that may exist in the vicinity of the first pixel) may belong to the same polygon region or are different from each other. may belong to and be adjacent to each other. Accordingly, the first polygon area and the second polygon area may be the same or different. Therefore, the adjacency matrix generating unit 33 should compare whether the first polygon area to which the first pixel belongs and the second polygon area to which the second pixel adjacent to the first pixel belongs are different from each other.

Then, the adjacency matrix generating unit 33 compares the first polygon area and the second polygon area with each other, determines whether the first polygon area and the second polygon area are different from each other according to the comparison result, and determines whether the first polygon area and the second polygon area are different from each other. If the region and the second polygon region are different from each other, a predetermined count value may be generated in response thereto, and an adjacency coefficient may be calculated based on the generated count value.

In addition, the adjacency matrix generating unit 33 includes a first polygon region to which a first pixel (ie, a pixel currently selected through raster scan) belongs and a second pixel adjacent to the first pixel (ie, a first pixel as an unselected pixel). When the comparison of the second polygon region to which the pixel belongs) is finished, the pixel adjacent to the right of the first pixel is updated as the first pixel, and the pixel adjacent to the right and adjacent to the bottom of the updated first pixel is completed. The comparison can be repeated by setting the pixel as the second pixel.

The adjacency matrix generator 33 may calculate a count value based on the number of second pixels belonging to the second polygon area determined to be different from the first polygon area.

For example, the adjacency matrix generating unit 33 reads labels of pixels to the right and bottom of the current pixel during raster scanning in the order of scanning, and determines whether there is a pixel different from the label of the current pixel. The adjacency matrix generator 33 may generate a predetermined count value (eg, +1) whenever there is a pixel different from the label of the current pixel, and calculate the adjacency coefficient using the generated count value. For example, if one of the right pixel and the lower pixel of the current pixel is different, the adjacency matrix generating unit 33 generates a count value of "+1" and calculates the adjacency coefficient by adding the generated count value to the adjacency coefficient. can If both the right pixel and the lower pixel of the current pixel are different, the adjacency matrix generating unit 33 generates a count value of "+2" and calculates the adjacency coefficient by adding the generated count value to the adjacency coefficient. have. Of course, if necessary, whenever the label of any one of the pixels to the right and the bottom of the current pixel is different from the label of the current pixel, it may be incremented by +2 or +3.

Referring to the region map of FIG. 6 , the first region r_1 and the second region r_2 are adjacent to each other, the second region r_2 and the third region r_3 are adjacent to each other, and the third region r_3 is adjacent to each other. and the fourth region r_4 are adjacent to each other, and the second region r_2 and the fourth region r_4 are adjacent to each other. In this case, the adjacency matrix for two different regions generated by the adjacency matrix generating unit 33 becomes M[r_1, r_2] = 6, M[r_2, r_3] = 2, M[r_3, r_4] = 3, M[r_2, r_4] = 5, and M[r_4, r_2] = 1. That is, by detecting whether the first region r_1 and the second region r_2 are adjacent to each other while scanning in the raster scan order, the label of the pixel 41 , 42 , 43 and the pixel to the right of the pixel 41 , 42 , 43 are detected. and labels of the lower pixels are different, so M[r_1, r_2] = 6. When scanning in the raster scan order and detecting whether the second region r_2 and the third region r_3 are adjacent to each other, the label of the pixels 51 and 52 and the label of the right pixel of the pixels 51 and 52 are different. [r_2, r_3] = 2. When scanning in the raster scan order and detecting whether the third region r_3 and the fourth region r_4 are adjacent to each other, the labels of the pixels 61 , 62 , 63 and the labels of the lower pixels of the pixels 61 , 62 , 63 are detected. is different, so M[r_3, r_4] = 3. When scanning in the raster scan order and detecting whether the second region r_2 and the fourth region r_4 are adjacent to each other, the label of the pixels 52 and 53 and the label of the lower pixel of the pixels 52 and 53 are different, and the pixel Since the label of (54, 55, 56) and the label of the right pixel of the pixel (54, 55, 56) are different, M[r_2, r_4] = 5. In addition, when scanning in the raster scan order and detecting whether the second region r_2 and the fourth region r_4 are adjacent to each other, the label of the pixel 71 and the label of the lower pixel of the pixel 71 are different, so M[r_4 , r_2] = 1. Accordingly, the adjacency matrix generating unit 33 determines whether the final adjacency matrix M as shown on the left of FIG. adjacency matrix including information) can be created.

The adjacency matrix generating unit 33 may apply symmetry to the adjacency matrix after raster scan is finished. Accordingly, an adjacency matrix as shown on the right side of FIG. 7 may be generated.

Meanwhile, in the above description, it is assumed that the calculated count value is +1, but if necessary, the calculated count value may be set differently depending on the size of the region. For example, if a large area (i.e. polygon) and a small area (i.e. polygon) are adjacent to each other, it would be desirable to merge the smaller area into the larger area. It is better to make the count value larger. Accordingly, a predetermined number of pixels are found each time the right and bottom pixels having a label different from the label of the current pixel are found, compared with other adjacent regions (ie, small regions) based on the pixels of the large area containing more than the reference number of pixels. It can also be increased by a weight value (eg, +3). In this way, if the same threshold value is given to each region, it may be preferable to count the larger region by a weight value.

As such, the adjacency matrix generating unit 33 can generate and update the adjacency matrix with one raster scan. Of course, if necessary, an initialized adjacency matrix M based on the number of adjacent two different polygon areas in the area map is first generated through raster scan, and then raster scan is performed again to pixels in the area map. For each of the pixels, whether the polygon region to which the corresponding pixel belongs is adjacent to another polygon region may be detected, and the detected result may be reflected in the initialized adjacency matrix.

When the raster scan is finally completed, it can be considered that the update of the adjacency matrix is finally completed. It can be transmitted to the merging condition generator 34 .

Accordingly, the merging condition generating unit 34 receives the adjacency matrix and then transmits a corresponding reception completion signal to the adjacency matrix generating unit 33 as an acknowledgment (ACK) signal.

Further, the merging condition generator 34 may generate (or set) a merging condition for merging between two adjacent different regions (ie, polygon regions). In this case, the merging condition generating unit 34 generates (or sets) a merging condition for merging between two adjacent regions (ie, polygon regions) based on the adjacency matrix generated by the adjacency matrix generating unit 33 . )can do.

Here, the merging condition may be a predetermined threshold.

In the embodiment of the present invention, when a raster scan is performed, it is possible to determine how many pixels in other areas are in contact with each area (ie, polygon area). That is, when the raster scan is finished, an adjacency matrix value for two different polygonal regions adjacent to each other is calculated. Accordingly, in the embodiment of the present invention, when the adjacency coefficient (that is, the adjacency matrix value) calculated by counting the adjacency of each pixel in the image with another area of the area to which the pixel belongs is greater than the threshold ( That is, when the threshold value is exceeded), the region to which the corresponding pixel belongs and other adjacent regions may be merged with each other. Here, the threshold value may be appropriately set based on adjacency matrix values for two different polygon regions adjacent to each other.

In this way, the merging condition generator 34 may generate a threshold value that may be a reference value for merging between two adjacent different regions (ie, polygon regions). Here, the threshold is updatable.

Additionally, the merging condition generating unit 34 may generate a merging direction when merging is performed.

For example, if it is assumed that two different regions adjacent to each other (ie, the first region and the second region) should be merged into one region, the merging direction means that the first region is merged into the second region, It may mean that the second area is merged into the first area. If the first area is to be merged into the second area, the values of all pixels in the first area will be changed to the area ID of the second area. Conversely, if the second region is to be merged into the first region, the values of all pixels in the second region will be changed to the region ID of the first region. This merging direction is for the convenience of processing, and if two different regions are to be merged into one region, it is considered that they may be merged in any direction.

In this way, the merging condition generating unit 34 may generate a merging direction when the threshold is satisfied.

On the other hand, the merging condition generator 34 may also use the size of the region (ie, polygon region) as an example of the merging condition.

In the previous description of the merging direction, the merging direction was described as being very free, but when the size of the region is used as an example of the merging condition, if there are two regions to be merged (that is, a polygon region), use the larger region as possible among the two regions. can be merged. Here, the larger area may mean an area in which the number of pixels included in the corresponding area is greater than that of other areas.

In the case of using the size of the polygon region as an example of a merging condition, the first polygon region having a predetermined size or more (eg, the number of pixels is 100 or more) and a size different from the first polygon region (eg, the number of pixels is about 5) When the adjacency matrix value between the second polygon regions is greater than the threshold, the first polygon region and the second polygon region will be merged with each other. However, when the first polygon area is quite large as described above, in order to prevent unnecessary merging due to the set threshold being too small, the threshold value may be readjusted so that the second polygon area is not merged with the first polygon area.

The merging condition generating unit 34 transmits a completion signal and a merging condition corresponding thereto to the merging map generating unit 35 as the generation of the merging condition is completed.

Of course, in the above description, the adjacency matrix generating unit 33 transmits the adjacency relation matrix to the merging condition generating unit 34. However, in the embodiment of the present invention, the merging condition is set in advance and the storage unit 20 ) can be pre-stored. Then, the merging condition generating unit 34 may not be needed, and the adjacency matrix generating unit 33 sends the adjacency matrix to the merging map generating unit 35 instead of sending it to the merging condition generating unit 34 . will be.

Accordingly, the merging map generator 35 receives the merging condition and transmits a reception completion signal corresponding thereto as an acknowledgment (ACK) signal to the merging condition generator 34 . Here, if the merging condition is previously stored, the merging map generator 35 reads the merging condition pre-stored in the storage unit 20 as the adjacency matrix is received.

Then, the merge map generating unit 35 generates a region to be merged based on the adjacency matrix generated by the adjacency matrix generating unit 34 (that is, a matrix indicating the degree of adjacency for each region in the region map). That is, it is possible to create a merge map representing the polygon area). For example, the merge map may be in the form of a look up table (LUT).

In other words, the merge map generator 35 may generate a merge map indicating the region ID to be merged based on the adjacent relationship matrix and the merge condition generated by the adjacent relationship matrix generator 34 .

For example, assuming that the threshold is 2, and taking an adjacency matrix for two different regions generated based on the region map of FIG. 6 as an example, the adjacency matrix having a value greater than the threshold is M[ r_1, r_2] = 6, M[r_2, r_4] = 5, and M[r_3, r_4] = 3. Accordingly, the merge map generator 35 merges the first region r_1, the second region r_2, and the fourth region r_4, and merges the third region r_3 and the fourth region r_4. A merge map can be created. In this case, the merge map generator 35 may generate a merge map such as LUT(r_1) = r_4, LUT(r_2) = r_4, and LUT(r_3) = r_4. Here, LUT(r_1) = r_4 means merging the first region into the fourth region, LUT(r_2) = r_4 means merging the second region into the fourth region, LUT(r_3) = r_4 means to merge the third region into the fourth region. Of course, if necessary, the merge map generator 35 instructs the first and fourth regions to be merged into the second region in consideration of the region (ie, polygon region) size (that is, LUT(r_1), respectively). = r_2, LUT(r_4) = r_2). Meanwhile, since the value of the adjacency matrix (ie, M[r_2, r_3]) of the second region r_2 with the third region r_3 is “2”, it is not greater than the threshold value, so the second region r_2 and the third region r_2 There is no need to merge the regions r_3 with each other. Therefore, when there is no need to merge, it is not necessary to generate a corresponding merge map or include it in the merge map.

As a result, the merge map generator 35 generates a look-up table (LUT) type merge map through a merge condition based on a nonzero component of the adjacency matrix M illustrated in FIG. 7 . it can be seen that In this case, the number of merge maps will be proportional to the number of polygon pairs adjacent to (or connected to) each other.

The merge map generator 35 transmits a corresponding completion signal and the merge map to the merge unit 36 as the merge map is generated.

Accordingly, the merging unit 36 receives the merging map and transmits a corresponding reception completion signal to the merging map generating unit 35 as an acknowledgment (ACK) signal.

In addition, the merging unit 36 may perform merging between two adjacent regions (ie, polygon regions) by applying the merging map to the region map received from the communication unit 10 . For example, when the first area is merged into the second area, the area ID of the pixel in the first area may be converted into the area ID of the pixel in the second area. The above-described merging may be performed between regions other than the first region and the second region.

Accordingly, the merging unit 36 may generate and output an area map in which two adjacent polygonal areas are merged. Here, the output area map may be transmitted to an external device.

8 is a flowchart illustrating an image processing method according to an embodiment of the present invention.

The image processing method according to an embodiment of the present invention is performed in the device 100 including the communication unit 10 , the storage unit 20 , and the control unit 30 .

The communication unit 10 of the device 100 receives the area map from the external device in response to the area map transmission signal from the external device, and outputs an area map reception completion signal as the reception of the area map is completed. Accordingly, the control unit 30 stores the area map in the storage unit 20 in response to the area map reception completion signal from the communication unit 10 . Also, a predetermined merging condition may be pre-stored in the storage unit 20 .

Then, the control unit 30 reads the stored area map (S10).

Next, the controller 30 generates an adjacency matrix for two adjacent different regions (ie, polygon regions) while raster-scanning the entire region map ( S12 ). In this case, since it is the first time to generate the adjacency matrix M for two adjacent regions (eg, r_1 and r_2), the value of the generated adjacency matrix M is initialized to “0 (zero)”. do.

In addition, in the process of scanning the area map in the raster scan order, for each of the pixels in the area map, the polygon area to which the current pixel (referred to as the first pixel) belongs and the second pixel adjacent to the current pixel belong The polygon areas are compared with each other, and the adjacency coefficient between the polygon area to which the current pixel (referred to as the first pixel) belongs and the polygon area to which the second pixel adjacent to the current pixel belongs is calculated according to the comparison result, and the calculated adjacency coefficient is used. to update the adjacency matrix M (S16). For example, it is determined whether there is a pixel different from the label of the current pixel by reading the labels of the pixels to the right and the bottom of the current pixel during scanning in the raster scan order. If there is a pixel different from the label of the current pixel, a predetermined count value (eg, +1) is generated, an adjacency coefficient is calculated based on the generated count value, and then the calculated adjacency coefficient is used in the adjacency matrix. increment the value. Here, incrementing may mean adding the count value to the current value of the corresponding adjacency matrix.

As such, when the raster scan is finally completed (“Yes” in S18), the adjacency matrix for two different adjacent regions existing in the region map is finally completed (that is, updated). For example, for the area map of FIG. 6 , an adjacency matrix as shown on the left in FIG. 7 may be finally completed (ie, updated).

Accordingly, the controller 30 generates a merge map indicating the area ID to be merged based on the finally completed adjacency matrix in the form of a look up table (LUT) (S20).

Finally, the controller 30 performs merging between two adjacent regions (ie, polygon regions) by applying the merge map to the region map received from the communication unit 10 ( S22 ). Accordingly, the controller 30 may generate and output an area map in which two adjacent polygonal areas are merged. Here, the output area map may be transmitted to an external device.

In addition, the above-described image processing method of the present invention can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. In addition, the computer-readable recording medium is distributed in a computer system connected through a network, so that the computer-readable code can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the method can be easily inferred by programmers in the art to which the present invention pertains.

The above description is merely illustrative of the technical spirit of the present invention, and various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be construed by the following claims, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

10: communication unit 20: storage unit
30: control unit 31: polygon generation unit
32: area map generation unit 33: adjacency matrix generation unit
34: merge condition setting unit 35: merge map generation unit
36: merging unit 100: image processing device

Claims (15)

a storage unit which stores in advance a region map including a region ID value for a polygon region to which each pixel of the image belongs; and
A processor having an arithmetic processing function; includes;
The processor by loading the instructions stored in the storage unit,
and generating an adjacency matrix including an adjacency coefficient representing adjacency between a plurality of polygon regions included in the image, wherein a first polygon region to which a first pixel of the pixels of the image belongs and an adjacency matrix adjacent to the first pixel The second polygon area to which the second pixel belongs is compared with each other, and it is determined whether the first polygon area and the second polygon area are different from each other according to the comparison result, and the first polygon area determined to be different from the first polygon area is determined. A count value is generated based on the number of second pixels adjacent to the first pixel among the second pixels belonging to the second polygon area, and a count value is generated between the first polygon area and the second polygon area based on the generated count value. calculating an adjacency coefficient, and generating the adjacency matrix using the calculated adjacency coefficient;
As the generation of the adjacency matrix is completed, polygon areas are merged based on the generated adjacency matrix to output a polygon area merged area map;
When merging the polygon regions, the processor compares an adjacency coefficient corresponding to the two adjacent polygon regions with a threshold value serving as a reference value for merging between the two adjacent polygon regions, and according to a result of the comparison merging the adjacent two different polygon regions when the coefficient exceeds a threshold,
image processing unit. delete The method of claim 1,
The processor is
When the comparison of the first polygon area to which the first pixel belongs and the second polygon area to which the second pixel adjacent to the first pixel belongs is finished, the pixel adjacent to the right of the first pixel is updated as the first pixel; repeating the comparison by setting the pixel adjacent to the right and the pixel adjacent below the updated first pixel as the second pixel;
image processing unit. delete The method of claim 1,
The storage unit further stores a merging condition for merging between two adjacent polygonal regions,
The processor merges polygon regions using the merging condition,
image processing unit. delete delete 6. The method of claim 5,
The merging condition includes a condition for the size of the polygon area,
The processor is
When merging the adjacent two different polygon areas, merging into a larger polygon area based on the size of the polygon area,
image processing unit. The method of claim 1,
The processor is
As the generation of the adjacency matrix is completed, a merge map indicating a polygon area to be merged is generated based on the generated adjacency matrix, and as the generation of the merge map is completed, the generated merged map is added to the area map. To create an area map where two adjacent polygonal areas are merged by applying,
image processing unit. The method of claim 1,
Further comprising; a communication unit for receiving the area map,
The communication unit sends the received area map to the processor and then sends an area map transmission completion signal to the processor;
the processor transmits an area map reception completion signal according to the reception completion of the area map as an ACK signal to the communication unit and then stores the area map in the storage unit;
image processing unit. A method performed in an image processing apparatus comprising a storage unit and a processor, the method comprising:
storing, by the storage unit, a region map including region ID values for polygon regions to which each pixel of the image belongs in advance;
wherein the processor generates an adjacency matrix including an adjacency coefficient representing an adjacency relationship between a plurality of polygon regions included in the image, wherein a first polygon region to which a first pixel of the pixels of the image belongs and the first polygon region A second polygon area to which a second pixel adjacent to one pixel belongs is compared with each other, and it is determined whether the first polygon area and the second polygon area are different from each other according to the comparison result, and is different from the first polygon area. A count value is generated based on the number of second pixels adjacent to the first pixel among the determined second pixels belonging to the second polygon area, and the first polygon area and the second pixel are generated based on the generated count value. calculating an adjacency coefficient between polygon regions and generating the adjacency matrix using the calculated adjacency coefficient; and
and outputting, by the processor, an area map in which the polygon areas are merged by merging polygon areas based on the generated adjacency matrix as the generation of the adjacency matrix is completed, and
The step of outputting the spirit map in which the polygon area is merged includes:
The adjacent coefficient corresponding to the two different adjacent polygon areas is compared with a threshold value that is a reference value for merging between the two adjacent different polygon areas, and if the adjacent coefficient exceeds the threshold according to the comparison result, the two adjacent different polygon areas are compared. to merge two polygonal areas,
Image processing method. delete 12. The method of claim 11,
The step of pre-saving the area map further stores a merging condition for merging between two adjacent polygon areas,
The step of outputting the area map in which the polygon areas are merged includes merging the polygon areas using the merging condition.
Image processing method. 12. The method of claim 11,
The step of outputting an area map in which the polygon area is merged includes:
As the generation of the adjacency matrix is completed, a merge map indicating a polygon area to be merged is generated based on the generated adjacency matrix, and as the generation of the merge map is completed, the generated merged map is added to the area map. To create an area map where two adjacent polygonal areas are merged by applying,
Image processing method. A computer-readable storage medium storing a program including instructions for performing the image processing method according to claim 11 .

Images