KR102548427B1 - Video Quality Measurement Apparatus and Method - Google Patents

Abstract

Embodiments of the present invention provide a video quality measuring apparatus and method capable of measuring the loss of encoded large-capacity video contents based on a unit structure included in a frame without using a high-performance measuring instrument.

Description

Video Quality Measurement Apparatus and Method {Video Quality Measurement Apparatus and Method}

An embodiment of the present invention relates to an apparatus and method for measuring image quality.

The information described below merely provides background information related to an embodiment according to the present invention and does not constitute prior art.

Thanks to the development of various electronic devices, it has become possible to create, process, deliver, and reproduce large-capacity video contents including ultra high definition (UHD) class (4K) video. Among various systems enabling the use of such large-capacity video content, the most representative system is an IPTV (internet protocol television) system.

An IPTV system provides a platform capable of complexly implementing services such as a live broadcasting service, video on-demand (VOD), and electronic commerce contents based on an Internet network. In a typical home, a separate external information receiving device such as a set-top box, cable box, hybrid box, IPTV box (internet protocol television box) serves as a medium to implement the IPTV system do

1 is a conceptual diagram illustrating an Internet environment in a typical home.

The home Internet environment includes an external information receiving device (110), a TV (120), a laptop PC (130), a wireless repeater (140), a mobile device (150), an IP network (160), a desktop PC (170), a game console ( 180), a network switch 190 and a content providing device 200.

The content providing device 200 encodes video content, converts it into a packetized stream (hereinafter, referred to as 'stream packet'), and transmits it to the external information receiving device 110. Encoded video content includes not only video but also audio information. The transmitted video content reaches the external information receiving device 110 via the IP network 160, the network switch 190, and the like.

The external information receiving device 110 receives encoded video content from the content providing device 200, and then decodes and outputs the video content to a display device.

A display device such as the TV 120 receives and outputs the decoded video content from the external information receiving device 110 . In addition to the TV 120, devices such as the laptop PC 130, the mobile device 150, and the desktop PC 170 may serve as display devices.

Data received from IP network 160 typically passes through network switch 190 first.

The network switch 190 is usually wired or wirelessly connected to the IP network 160, receives data from the outside, and serves to switch them to various electronic devices.

Meanwhile, video content is encoded using a standard codec or a non-standard codec. In the case of encoding using a codec, the content providing device 200 includes an I frame (intra coded frame), a P frame (predictive coded frame), a B frame (bipredictive coded frame), and a D frame ( A frame such as DC direct coded frame) can be used.

Packet loss may occur during transmission of an encoded video. If packet loss occurs in an I frame, most of the playback picture may be lost. If packet loss occurs in an I frame, it is known that the loss of an I frame greatly affects user experience quality because a screen being reproduced may be stopped. In addition, even when packet loss occurs in a portion adjacent to an I frame, most of a reproduced screen may be lost, which may greatly affect user perceived quality.

Therefore, in order to manage video quality, it is necessary to detect frame loss, measure video quality, and prepare management standards for this.

Conventionally, image quality was measured using a high-performance measuring instrument. Such a measuring instrument receives encoded video stream packets from an IP network, demultiplexes them, decodes the demultiplexed video signals, and measures quality in frame units.

However, this method has a disadvantage in that a separate device must be installed. In addition, in the case of large-capacity video content, even if HEVC, which has good compression efficiency, is used, computational complexity is very high, so a large amount of computation and memory are required to decode all frames and detect frame loss. Such a large amount of computation may affect existing services provided by an external information receiving device such as a set top box, and may degrade performance of the external information receiving device.

Therefore, there is a need for an apparatus and method capable of measuring frame loss of large-capacity video contents of UHD level (4K) or higher without degrading the performance of an external information receiving apparatus by minimizing the amount of computation required for decoding.

The main object of the embodiments of the present invention is to provide an image quality measurement apparatus and method capable of measuring loss of a unit structure included in an image frame without using a high-performance measuring instrument.

According to an embodiment of the present invention, a transmission and reception unit for receiving a video stream packet divided into a plurality of frames and encoded from a content providing apparatus; a damaged unit structure information generator configured to generate damaged unit structure information by detecting at least one damaged unit structure among at least one unit structure included in the frame; a reference relationship analyzer for determining a damage propagation frame, which is a frame that directly or indirectly refers to the damaged frame including the damaged unit structure; and a user perceived quality index calculator calculating a user experience quality index based on the number of damaged frames and damaged propagation frames.

According to an embodiment of the present invention, in a method for measuring video quality using a video quality measurement apparatus, a video stream packet divided into a plurality of frames and encoded is received from a content providing apparatus. receiving process; generating damaged unit structure information by detecting at least one damaged unit structure among at least one unit structure included in the frame; identifying a damage propagation frame, which is a frame that directly or indirectly refers to a damaged frame including the damaged unit structure; and calculating a user experience quality index based on the number of the damaged frames and the damaged propagation frames.

According to an embodiment of the present invention, there is an effect of enabling loss of encoded large-capacity video content to be measured based on a unit structure included in a frame without using a high-performance measuring instrument.

According to another aspect of an embodiment of the present invention, there is an effect of enabling unit structure loss of a UHD (ultra high definition) class (4K) channel to be measured without affecting the existing service provided by the external information receiving apparatus.

According to another aspect of an embodiment of the present invention, since a damaged frame is processed based on a actually displayed frame, there is an effect of reflecting the subjective quality recognized by the user well.

1 is a conceptual diagram illustrating an Internet environment in a typical home.
2 is a conceptual diagram of an Internet environment to which an apparatus for measuring image quality according to an embodiment of the present invention is applied.
3 is a schematic configuration diagram of an apparatus for measuring image quality according to an embodiment of the present invention.
4 is a diagram for explaining propagation of corruption between frames in an encoded video stream packet.
5 is a diagram for explaining damage propagation between unit structures in an encoded video stream packet.
6 is a diagram for explaining corruption propagation between frames in a multi-view service.
7 is a diagram illustrating a frame display order, a frame decoding order, and a reference relationship between frames in an encoded video stream packet.
8 is an exemplary diagram for grasping GOP structure information in a video quality measurement method according to an embodiment of the present invention.
9 is a flowchart illustrating a method for measuring image quality according to an embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in adding reference numerals to components of each drawing, the same components have the same numerals as much as possible, even if they are displayed on different drawings. In addition, in describing an embodiment of the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of an embodiment of the present invention, the detailed description will be omitted.

Hereinafter, a video quality measurement apparatus and method according to an embodiment of the present invention will be described with reference to the accompanying drawings.

2 is a conceptual diagram of an Internet environment to which an apparatus for measuring image quality according to an embodiment of the present invention is applied.

The content providing device 200 according to an embodiment of the present invention encodes video content, converts it into a packetized stream (hereinafter referred to as 'stream packet'), and transmits the video content to the IP network 160.

The content providing device 200 uses a codec based on the MPEG-2 standard (ISO/IEC13818), H.264/AVC (advanced video coding), H.265/HEVC (high efficiency video coding), etc. Content can be encoded.

When encoding large-capacity video content, the content providing apparatus 200 uses some or all of I frames, P frames, B frames, and D frames to divide them into a plurality of frame units and generate video stream packets.

Therefore, in order to manage the user's quality of experience (QoS) for video content received by the external information receiving device 110, particularly large-capacity video content, it is necessary to recognize the frame type.

The video quality measuring device 300 according to an embodiment of the present invention decodes the video stream packet received from the content providing device 200 and transmits it to the TV 120, the network switch 190 and the TV 120 located between

The image quality measuring device 300 is manufactured in the form of a small module and can be embedded in the external information receiving device 110 as well as the laptop PC 130, the desktop PC 170, and the mobile device 150. Also, the video quality measuring device 300 may be integrated into a TV 120 such as a smart television (Smart TV).

3 is a schematic configuration diagram of an apparatus for measuring image quality according to an embodiment of the present invention.

Before receiving a video stream packet and decoding the received packet, the video quality measurement apparatus 300 according to an embodiment of the present invention detects a damage unit structure in the video stream, and frames included in the video stream packet. By figuring out the reference relationship between them, the range of video stream damage is quantified objectively.

An image quality measuring device 300 according to an embodiment of the present invention includes a transceiver 310, a damage unit structure information generator 320, a reference relationship analyzer 330, and a user perceived quality index calculator 340. include

Although not shown in FIG. 3, those of ordinary skill in the technical field of the present invention, in addition to the components for measuring the image quality shown in FIG. It will be appreciated that it may further include components necessary to perform.

The transceiver 310 receives the video stream packet from the content providing device 200 and transmits the calculated video quality information to the network manager or network management device. The transceiver 310 may use a transmitting control protocol (TCP) and a user datagram protocol (UDP) together with IP.

The corruption unit structure information generation unit 320 detects a frame including a corrupted unit structure in the video stream packet before the encoded video stream packet is completely decoded, and generates the corruption unit structure information.

One frame includes at least one unit structure such as a tile or a slice. Damaged unit structure means the unit structure in which loss or damage first occurred. The damaged frame refers to a frame including at least one damaged unit structure. A reference frame refers to a frame that refers to information of at least one other frame in order to form its own frame information. That is, a frame using information of at least one other frame to form information of one frame is a reference frame.

The reference relationship analyzer 330 generates a frame reference relationship that is a reference relationship between frames included in the video stream packet.

The reference relationship analysis unit 330 analyzes headers of frames around the damaged frame to generate a frame reference relationship. The reference relation analyzer 330 may parse frames around the damaged frame even when determining the period (intra period) of the I frame or the size of the GOP structure.

Header information of a frame includes a reference picture set (RPS), a picture order count (POC), a video parameter set (VPS), a sequence parameter set (SPS), a picture parameter set (PPS), and the like.

Here, if the encoded video stream packet is encoded based on HEVC, the parsed header may not be a frame header but a slice header.

If the content providing device 200 and the video quality measurement device 300 share a GOP structure, the reference relationship analyzer 330 may not perform an additional header analysis task to determine the GOP structure.

Also, the reference relationship analyzer 330 may determine the GOP structure using network abstraction layer (NAL) header information instead of header information of frames. NAL header information includes temporal_id information. The temporal_id information provides information on a temporal layer of a video stream packet.

The reference relationship analyzer 330 may determine the I frame period or GOP structure in the video stream packet by using the value of temporal_id of 0.

For distinction from a reference frame, assume that a frame that does not refer to any other frame is an independent frame. A frame reference relationship, which is a reference relationship between frames, is a relationship between at least two frames. That is, the reference frame means that information of at least one reference frame or independent frame is used to form its own information. Also, that at least two frames are in a reference relationship means that at least one frame among the at least two frames is a reference frame.

The reference relationship analyzer 330 identifies an indirectly damaged frame, which is a frame affected by the damaged frame in a primary, secondary, tertiary or higher order, using the frame reference relationship and damaged frame information. That is, a frame affected by propagation of damage or loss from the found damaged frame is identified, and indirect damaged frame information, which is information on the indirectly damaged frame, is generated.

The reference relation analysis unit 330 determines frames that directly or indirectly refer to a damaged frame including at least one damage unit structure.

An image quality measurement apparatus according to an embodiment of the present invention may include a damaged frame processing unit (not shown) that performs damage processing to discard damaged frames and damaged propagation frames that are frames that directly or indirectly refer to the damaged frames. there is.

Meanwhile, in HEVC, the reference relationship may be much more flexible than in H.264/AVC or MPEG-2 standards (ISO/IEC13818). In other words, in terms of an encoding device, more frames can be used as reference frames to generate one frame, and the degree of freedom of the encoding order for the display order is also much higher.

Here, the encoding order means the order in which each frame is encoded in the video stream.

In HEVC, a B frame may be used as a reference frame to encode another B or P frame. This improves the compression efficiency of HEVC. However, it has the disadvantage of causing error propagation if some data is lost or corrupted.

A very useful structure used in HEVC is the hierarchy of B frame or hierarchical B. Because the B frame hierarchy strictly limits the number of frames affected by a particular data corruption, it can provide very high compression efficiency while limiting error propagation.

In general, the more I-frames a video stream has, the more freely it can be encoded and modified. However, there is a disadvantage in that the bit rate required for encoding the video increases as the video stream has more I frames.

The user perceived quality index calculation unit 340 is a user perceived quality index (user experience quality index) based on the damage unit structure information obtained from the damage unit structure information generation unit 320 and the frame reference relationship obtained from the reference relationship analysis unit 330. ) is calculated.

The user-perceived quality indicator calculation unit 340 uses the number of frames that directly or indirectly refer to damaged frames including at least one damaged unit structure to calculate the user-perceived quality indicator.

In addition, the user-perceived quality indicator calculation unit 340 calculates the user-perceived quality indicator, the maximum damaged area of the damaged frame, the number of damaged frames included in a plurality of divided frame units, and included in a plurality of divided frame units. Calculate the number of all frames. The maximum damaged area of the damaged frame can be obtained by determining the number of at least one damaged unit structure included in the damaged frame.

In addition, the video quality measuring apparatus according to an embodiment of the present invention may not separately process damaged frames and unit structures included in the damaged frames.

4 is a diagram for explaining propagation of corruption between frames in an encoded video stream packet.

An encoded video stream 430 according to an embodiment of the present invention shown in FIG. 4 includes a plurality of GOPs 420, and its structure is a B frame hierarchy.

One GOP 420 includes eight frames 410 . A frame marked with a filled circle represents a damage occurrence frame 412 in which damage occurs, and a frame marked with an empty circle represents a damage propagation frame 414 that is indirectly damaged by the damage occurrence frame 412 .

(a) of FIG. 4 shows a case where the damage occurrence frame 412 exists at frame position 0. In this case, frames existing at frame position 1, frame position 2, and frame position 8 are affected by the primary reference relationship.

Frames present at frame position 3, frame position 4, frame position 6, and frame position 7 due to reference relationships by the frames present at frame position 1, frame position 2, and frame position 8 affected by the primary reference relationship. are affected

Similarly, the frame at frame position 5 is affected by the tertiary reference relationship of the frames at frame position 3, frame position 4, frame position 6 and frame position 7 affected by the secondary reference relationship, As a result, all frames belonging to one GOP become corrupted propagation frames 414 .

(b) of FIG. 4 is a case where the damage occurrence frame 412 exists at the frame position 1. In this case, since no frame refers to the damaged frame 412 existing at frame position 1, only the damaged frame 412 existing at frame position 1 becomes a damaged frame.

(c) of FIG. 4 shows a case where the damaged frame 412 exists at frame position 2. In this case, frames existing at frame position 1 and frame position 3 are affected by the primary reference relationship. However, since no frames refer to the frames at frame position 1 and frame position 3, the frames at frame position 2, which are the frames where the corruption occurred, and the frames at frame position 1 and frame position 3 affected by the primary reference relationship are the damaged propagation frames. (414).

(d) of FIG. 4 shows a case where the damage occurrence frame 412 exists at frame position 3. In this case, since no frame refers to the damaged frame 412 existing at frame position 3, only the damaged frame 412 existing at frame position 3 becomes a damaged frame.

(e) of FIG. 4 shows a case where the damage occurrence frame 412 exists at frame position 4. In this case, frames existing at frame positions 2, 3, 5, and 6 are affected by the primary reference relationship to become the damage propagation frame 414.

Among the frames existing at frame positions 2, 3, 5, and 6 affected by the primary reference relationship, no frames refer to frame positions 3 and 5.

Accordingly, the frames present at frame positions 1 and 7 are affected by the frames present at frame positions 2 and 6, resulting in the damage propagation frame 414. That is, in this case, all frames other than frame position 0 and frame position 8 are affected and become the damage propagation frame 414 .

(f) of FIG. 4 shows a case where the damage occurrence frame 412 exists at frame position 5.

In (f) of FIG. 4 , since no frame refers to the damaged frame 412 existing at the frame position 5, only the damaged frame 412 existing at the frame position 5 becomes a damaged frame.

(g) of FIG. 4 shows a case where the damage occurrence frame 412 exists at the frame position 6.

In (g) of FIG. 4 , frames existing at frame positions 5 and 7 are affected by the primary reference relationship. However, since there are no frames referencing the frames at frame positions 5 and 7, the frame at frame position 6, which is the frame where the damage occurred, and the frame at frame position 5 and 7 affected by the primary reference relationship. Frame damage becomes a frame.

(h) of FIG. 4 is a case where the damage occurrence frame 412 exists at frame position 7. In this case, since no frame refers to the damaged frame 412 existing at frame position 7, only the damaged frame 412 existing at frame position 7 becomes a damaged frame.

(i) of FIG. 4 shows a case where the damage occurrence frame 412 exists at the frame position 8. In this case, frames existing at frame positions 4, 6, and 7 are affected by the primary reference relationship.

Frames present at frame position 2, frame position 3 and frame position 5 are affected due to the reference relationship by which the frames present at frame position 4, frame position 6 and frame position 7 are affected by the primary reference relationship. .

Similarly, the frame existing at frame position 1 is affected by the tertiary reference relationship of the frames existing at frame position 2, frame position 3, and frame position 5 that are affected by the secondary reference relationship, so that frame position 0 is affected. Frames existing in all frame positions except for the damage propagation frame 414.

5 is a diagram for explaining damage propagation between unit structures in an encoded video stream packet.

5(a) shows damage propagation between unit structures when the unit structure is a frame. When the unit structure is a frame, one unit structure forms one frame, and in this case, the frame referring to the damaged frame 512 becomes the damage propagation frame 514.

5(b) shows damage propagation between unit structures when the unit structure is a tile. When the unit structure is a tile, a plurality of tiles form one frame. The damaged tile 522, which is a damaged tile, propagates damage to tiles at the same location included in the reference frame by reference relationship. When the damage unit structure is a tile, damage caused by the damage occurrence tile 522 is limited to only the tile.

The total number of tiles included in the damaged frame is 24, and the number of damaged tiles 522 is 2. Referring to this frame, the frame to which damage is propagated also includes 24 tiles, and in this case, the number of damage propagation tiles 524 is also 2.

5(c) shows damage propagation between unit structures when the unit structure is a slice. When the unit structure is a slice, a plurality of slices form one frame. Corrupted slice 532, which is the slice where the corruption occurred, propagates the corruption to the slice at the same location included in the reference frame. When the damage unit structure is a slice, damage caused by the damage occurrence slice 532 is limited to only the slice.

The total number of slices included in the damaged frame is 4, and the number of damaged slices 532 is 1. Referring to this frame, the frame to which corruption is propagated also includes 4 slices, and in this case, the number of corruption propagation slices 534 is 1.

The user-perceived quality indicator calculator 340 may calculate the user-perceived quality indicator using Equations 1, 2, and 3.

Here, FCI, NF _C,GOP , NF _T,GOP , NF _{C,I -frame} and NF _{T,I -frame} are the damage index per frame, the number of damaged frames included in one GOP, and the number of damaged frames included in one GOP, respectively. The total number of frames, the number of damaged frames included in one I frame period, and the total number of frames included in one I frame period.

Where, IFCI, A _C , _AF , NT _CT , NT _TT , NT _CS , and NT _TS are the intra-frame damage index, the damage area within the frame, the number of damaged tiles within the frame, the total number of tiles within the frame, and the number of damaged slices within the frame, respectively. and the total number of slices in the frame.

Here, UEQI is a user perceived quality index.

For example, the GOP size is 8, the total number of tiles in a frame is 24, the damaged frame is frame position 2, the frames referencing this damaged frame are two frames at frame position 1 and frame position 3, and the damaged tile is, Assume that the tiles exist at tile position 7 and tile position 15.

The number of damaged frames included in one GOP is 3 as the sum of the first damaged frame and two frames referencing this frame. The damaged area in the frame equals the number of damaged tiles in the frame, so it is 2. Accordingly, it can be seen that the user-perceived quality index is 0.003125 as (3÷8)×(2÷24).

6 is a diagram for explaining corruption propagation between frames in a multi-view service.

The N multi-view frame 610 and the N+α multi-view frame 620 include four channels (Channel A, Channel B, Channel C, and Channel D). When damage occurs in channel D, damage also occurs in channel D 614 in the N+α multiview frame 620, which is a frame referring to the frame 610 including the damaged channel.

7 is a diagram illustrating a frame display order, a frame decoding order, and a reference relationship between frames in an encoded video stream packet.

Referring to FIG. 7 , an intra period, that is, a period of an I frame is 30 as the number of frames between an I0 frame and an I1 frame. The display order is I0 frame, B2 frame, B1 frame, B3 frame, B0 frame, B5 frame, B4 frame, B6 frame, P0 frame, B9 frame, B8 frame, B10 frame, B7 frame, B12 frame, B11 frame, B13 frame. frame, P1 frame, ... … am. The decoding order is I0 frame, P0 frame, B0 frame, B1 frame, B2 frame, B3 frame, B4 frame, B5 frame, B6 frame, P1 frame, B7 frame, B8 frame, B9 frame, B10 frame, B11 frame, B12 frame. , B13 frame, … … am.

Referring to the display order, the B0 frame refers to the I0 frame and the B0 frame. The B1 frame refers to the I0 frame and the B0 frame, and the B2 frame refers to the I0 frame and the B1 frame. The B3 frame refers to the B0 frame and the B1 frame, and the B4 frame refers to the B0 frame and the P0 frame. The B5 frame refers to the B0 frame and the B4 frame, and the B6 frame refers to the B4 frame and the P0 frame.

There are no frames that reference the B2 frame, the B3 frame, the B5 frame, and the B6 frame. In this way, the reference relationship can also be grasped for the frames after the B7 frame.

On the other hand, the size (period) of the GOP is 8 from the B2 frame to the P0 frame or from the B9 frame to the P1 frame.

The display order within the GOP is also the same as the display order within the entire intra period. The reference relationship from the B2 frame to the P0 frame is the same as that within the intra period. That is, the B0 frame refers to the I0 frame and the B0 frame. The B1 frame refers to the I0 frame and the B0 frame, and the B2 frame refers to the I0 frame and the B1 frame. The B3 frame refers to the B0 frame and the B1 frame, and the B4 frame refers to the B0 frame and the P0 frame. The B5 frame refers to the B0 frame and the B4 frame, and the B6 frame refers to the B4 frame and the P0 frame. There are no frames that reference the B2 frame, B3 frame, B5 frame, and B6 frame.

If the video quality measuring device 300 or the external information receiving device 110 such as a set-top box does not know the structure of the GOP, in order to determine the structure of the GOP, the headers of the frames around the damaged frame can be identified analyze until That is, assuming that the damaged frame is the B1 frame, headers of all frames from the B2 frame located after the B1 frame in the decoding order until the GOP structure is identified must be analyzed in order to determine the GOP structure.

8 is an exemplary diagram for grasping GOP structure information in a video quality measurement method according to an embodiment of the present invention.

This method is a method that can be used when network abstract layer unit (NALU) header information can be identified. The NALU header includes information such as forbidden_bit, nal_unit_type, and temporal_id. Referring to FIG. 8 , it can be seen that frames having temporal_id of 0 are I0 frames, P0 frames, and P1 frames. Frames with a temporal_id of 1 are B0 frames and B7 frames, and frames with temporal_id of 2 are B1 frames, B4 frames, B8 frames, and B11 frames.

The reference relationship analyzer 330 may determine the intra period and GOP structure through the value of temporal_id equal to 0. When the GOP structure can be identified using the NALU header information, there is an advantage in that the headers of the frames do not need to be separately analyzed to determine the GOP structure.

9 is a flowchart illustrating a method for measuring image quality according to an embodiment of the present invention.

An image quality measuring method according to an embodiment of the present invention measures image quality by determining the number of damaged frames including damaged unit structures and damaged propagation frames that directly or indirectly refer to the damaged frames.

For example, assuming that frame B1 is lost, the transceiver 310 receives an encoded video stream packet from the content providing device 200 (S910). The encoded video stream packet may pass through the IP network 160 and be transferred through the in-house network switch 190 or the wireless repeater 140.

The damage unit structure information generation unit 320 detects at least one unit structure in which damage has occurred among at least one unit structure included in the frame, and generates damage unit structure information (S920). Assuming that the first slice of the B1 frame is damaged, information on the first slice and information on the frame including the first slice become damage unit structure information.

The reference relation analysis unit 330 identifies frames that directly or indirectly refer to the damaged frame including at least one damaged unit structure (S730). Since the B2 and B3 frames refer to the B1 frame, the B2 and B3 frames are corruption propagation frames.

The user-perceived quality indicator calculator 340 calculates a user-perceived quality indicator based on the damaged frame and the frame reference relationship (S940).

The user-perceived quality indicator calculation unit 340 may calculate the user-perceived quality indicator by confirming that the B2 frame and the B3 frame refer to the B1 frame, which is a damaged frame.

The user-perceived quality indicator calculator 340 may transmit the calculated result to an external network manager or network management device (not shown) through the transceiver 310 .

Also, the user-perceived quality indicator calculator 340 may accumulate the user-perceived quality indicators for a predetermined target period and calculate statistics including an average, maximum value, variance, or standard deviation for the user-perceived quality indicators.

The method for measuring video quality according to an embodiment of the present invention can measure video quality using the same method as the above-described method for video stream packets capable of providing a multi-view service.

In the multi-view service, each view frame is regarded as one tile, and one multi-view frame is regarded as composed of a plurality of tiles. Therefore, if even one of the plurality of views is damaged, it is assumed that the frames constituting the entire multi-view screen are damaged, and the image quality can be measured in the same way as the above method.

In FIG. 9, it is described that each process is sequentially executed, but is not necessarily limited thereto. In other words, since the process described in FIG. 9 may be changed and executed or one or more processes may be applied in parallel, FIG. 9 is not limited to a time-series sequence.

Meanwhile, each step of the flowchart shown in FIG. 9 can be implemented as computer-readable codes on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. That is, computer-readable recording media include magnetic storage media (eg, ROM, floppy disk, hard disk, etc.), optical reading media (eg, CD-ROM, DVD, etc.) and carrier waves (eg, Internet Transmission through) and the same storage medium. In addition, computer-readable recording media may be distributed to computer systems connected through a network to store and execute computer-readable codes in a distributed manner.

The above description is only illustrative of the technical idea of one embodiment according to the present invention, and those skilled in the art to which one embodiment according to the present invention belongs can variously Modifications and variations will be possible. Therefore, the embodiments according to the present invention are not intended to limit the technical idea of this embodiment, but to explain, and the scope of the technical idea of this embodiment is not limited by these embodiments. The scope of protection of one embodiment according to the present invention should be construed according to the claims below, and all technical ideas within the equivalent range should be construed as being included in the scope of rights of one embodiment according to the present invention.

An embodiment of the present invention is applied to the technical field of measuring video quality in an IPTV system, etc., so that the loss of a unit structure included in a video frame can be measured without using a high-performance measuring instrument, thereby reducing cost. It is a useful invention.

100: Internet environment 110: External information receiving device
120: TV 130: laptop PC
140: wireless repeater 150: mobile device
160: Internet 170: Desktop PC
180: game console 190: network switch
300: video quality measuring device 310: transceiver
320: Damage frame information generation unit 330: Reference relationship analysis unit
340: user perception quality indicator calculation unit 410: frame
412, 512: damage occurrence frame 414, 514: damage propagation frame
522 Corruption occurrence tile 524 Corruption propagation tile
532 damage occurrence slice 534 damage propagation slice
610: N multi-view frame 620: N+α multi-view frame
612: damage generation channel 614: damage propagation channel

Claims (10)

a transmission/reception unit for receiving a video stream packet divided into a plurality of frames and encoded from a content providing device;
a damaged unit structure information generator configured to generate damaged unit structure information by detecting at least one damaged unit structure among at least one unit structure included in the frame;
a reference relationship analyzer for determining a plurality of frame units to which the damaged unit structure belongs, and determining a damage propagation frame, which is a frame that directly or indirectly refers to a damaged frame including the damaged unit structure within the plurality of frame units; and
A user perceived quality index calculator calculating a user experience quality index based on the number of damaged frames and damaged propagation frames.
Including,
The plurality of frame units,
It is classified based on the size of the I frame (intra coded frame) period or GOP (group of pictures) structure,
The reference relationship analysis unit,
Until identification of a plurality of frame units to which the damaged unit structure belongs is completed, the damaged propagation frames are identified by parsing headers of frames following the damaged frame including the damaged unit structure in decoding order,
The user perceived quality indicator calculation unit,
The video quality measurement apparatus characterized in that a value obtained by adding the number of damaged frames and the number of damaged propagation frames is used as a user-perceived quality indicator. delete delete According to claim 1,
The user perceived quality indicator calculation unit,
Image quality measuring device, characterized in that for calculating the maximum damage area that can occur in the damaged frame. According to claim 4,
The user perceived quality indicator calculation unit,
The image quality measuring device characterized in that the number of damaged unit structures is used to obtain the maximum damaged area. delete According to claim 4,
The user perceived quality indicator calculation unit,
The image quality measurement apparatus according to claim 1 , wherein a value obtained by multiplying a value obtained by adding the number of damaged frames and the number of damaged propagation frames multiplied by the maximum damaged area is used as a user-perceived quality indicator. According to claim 7,
An image quality measuring device characterized in that it does not separately process damage within a frame or damage in a frame unit. A method for measuring image quality using an image quality measurement device,
Receiving a video stream packet divided into a plurality of frames and encoded from a content providing device;
generating damaged unit structure information by detecting at least one damaged unit structure among at least one unit structure included in the frame;
identifying a plurality of frame units to which the damaged unit structure belongs, and identifying a damage propagation frame that is a frame that directly or indirectly references a damaged frame including the damaged unit structure within the plurality of frame units; and
Calculating a user experience quality index based on the damaged frame and the number of damaged propagation frames
Including,
The plurality of frame units,
It is classified based on the size of the I frame (intra coded frame) period or GOP (group of pictures) structure,
The process of identifying the damage propagation frame,
Until identification of a plurality of frame units to which the damaged unit structure belongs is completed, the damaged propagation frame is identified by parsing headers of frames following the damaged frame including the damaged unit structure in decoding order,
The process of calculating the user perceived quality index,
A method for measuring video quality characterized in that a value obtained by adding the number of damaged frames and the number of damaged propagation frames is used as a user-perceived quality indicator. According to claim 9,
The process of calculating the user perceived quality index,
The method of measuring image quality, characterized in that by accumulating the user-perceived quality indicators for a predetermined target period, statistics including an average, maximum value, variance or standard deviation for the user-perceived quality indicators are calculated.

Images