TWI502998B - Method and system of video information acquisition based on video data traffic - Google Patents

Description

Video information acquisition method and system based on video data traffic

The present invention relates to a method and system for obtaining video information, and more particularly to a method and system for obtaining video information based on video data traffic.

With the rapid development of Internet-related technologies, more and more multimedia information is transmitted over the Internet, such as pictures, video streams, and so on. However, in the case of limited network bandwidth, excessive network packets may cause packet delay or packet loss, causing the receiver to take more time to receive completely. When the network packet is received, even the complete network packet cannot be received, which also causes the user to watch the video through the Internet, and the video picture may be inconsistent or broken. Therefore, how to efficiently transmit network packets becomes an important issue.

One of the existing methods for solving the above problem is to determine the picture type of the video image by the application, so that the network packet is transmitted on the Internet in a more efficient manner. This is done by adding a Real Time Protocol Header (RTP Header) to each network packet. The instant protocol header includes a data field for representing the picture type corresponding to the image frame. This data field is used by the application on the transmission side to fill in the information of the screen type. Therefore, when the network packet is transmitted, the data field can be used by other network endpoint applications to determine the picture type of the image frame corresponding to the network packet.

The shortcomings of the above methods are two. First, for different image standards (such as H.261, MPEG4, etc.), the instant protocol headers will have different definitions, and the application on the transmission side may not completely fill in the picture type. Information; second, in addition to the transmitting end, other network endpoints in the transmission process must have the ability to resolve network packets with instant protocol headers.

Accordingly, it is an object of the present invention to provide a method of obtaining video information based on video data traffic.

Therefore, the method for obtaining video information based on video data traffic of the present invention comprises the following steps.

(A) receiving an image picture packet, and recording a receiving time and a packet data flow corresponding to the image picture packet. (B) Calculating a packet interval time based on the reception time and a previous reception time of the previous video picture packet corresponding to the video picture packet. (C) Comparing the packet interval time with an average picture interval time.

(D) If the packet interval time is less than or equal to the average picture interval time, an accumulated data flow rate is obtained according to the packet data flow, and the process returns to step (A). (E) If the packet interval time is greater than the average picture interval time, the accumulated data traffic at this time is used as a picture data flow of an image frame. (F) determining a corresponding picture type of the image picture based on the data flow rate of the picture.

Another object of the present invention is to provide a video information obtaining system based on video data traffic.

Therefore, the video information obtaining system based on the video data traffic of the present invention comprises a receiving unit, a packet information obtaining unit, and a picture type analyzing unit.

The receiving unit is configured to receive a plurality of image frame packets.

The packet information obtaining unit is configured to calculate a frame data traffic according to the image frame packets, wherein for each image frame packet, the packet information obtaining unit is configured to perform the following actions: recording a receiving time corresponding to the image frame packet And a packet of data traffic.

A packet interval time is calculated according to the receiving times of the corresponding image frame packets and the adjacent image frames.

A cumulative data flow is obtained based on the packet data flow.

According to the packet interval time, it is determined whether the accumulated data traffic is used as a picture data flow of an image frame.

The picture type analyzing unit is configured to determine a picture type of the image picture according to the picture data flow.

The effect of the present invention is that no matter whether the application fills in the information of the screen type or whether the endpoints in the transmission process have the ability to parse the network packet with the instant protocol header, the data traffic through the screen is used. , the picture type of the image picture can be judged to obtain video information.

The above and other technical contents, features and advantages of the present invention will be apparent from the following detailed description of the preferred embodiments.

Referring to FIG. 1, a preferred embodiment of the video information obtaining system 1 based on video data traffic of the present invention comprises a receiving unit 11, a packet information obtaining unit 12, and a frame type analyzing unit 13. In the preferred embodiment, the packet information obtaining unit 12 and the picture type analyzing unit 13 of the video information obtaining system 1 based on the video data traffic are implemented in a software manner, and the implementation manner is the internal storage based on the video data traffic. The video information obtained by the program is a computer program product. When an electronic device (for example, a processor of a computer, not shown) is loaded into the program and executed, the packet information obtaining unit 12 and the function of the picture type analyzing unit 13 can be completed.

The receiving unit 11 is configured to receive a plurality of image frame packets.

The packet information obtaining unit 12 is configured to calculate a frame data traffic according to the image frame packets, wherein for each image frame packet, the packet information obtaining unit 12 is configured to perform the following actions: recording one corresponding to the image frame packet Receiving time and a packet data flow; calculating a packet interval time according to the receiving time corresponding to the image frame packet and the adjacent image frame packet; and obtaining a cumulative data flow according to the packet data flow; The packet interval time determines whether the accumulated data traffic is used as a picture data flow of an image frame.

The picture type analyzing unit 13 is configured to determine a picture type of the picture picture according to the picture data flow.

Therefore, the video processing related application installed in the computer (not shown) can perform subsequent processing on the corresponding image frame packet according to the picture type of the image picture, so that the image picture package can be more efficiently Being transmitted. The processing performed by the application related to the video processing for the video picture packet is easily understood by those skilled in the art and is not the focus of the present invention, and therefore will not be described herein.

Referring to FIG. 1 and FIG. 2, in the following, in accordance with the method for obtaining video information based on video data traffic according to the present invention, the steps of operation between the above units in the preferred embodiment are further described.

As shown in step S0, the packet information obtaining unit 12 initializes, that is, initializes a count value and a cumulative data flow to zero; and sets an average picture interval time and an average data flow rate as a preset average. Screen interval time and a preset average data flow. The preset average picture interval time and the preset average data flow are experience values; the experience values are calculated according to a plurality of packet interval times calculated by a receiving time of a plurality of network packets, and are calculated by a plurality of network packets. The average of the multiple screen data flows is calculated.

As shown in step S1, the receiving unit 11 first receives an image picture packet.

As shown in step S2, the packet information obtaining unit 12 records the receiving time and the packet data traffic corresponding to the image frame packet, and encapsulates the previous (Previous) image frame corresponding to the image frame packet according to the receiving time. A packet interval is calculated based on a previous reception time.

That is, when the image frame packet is the first image frame packet received by the receiving unit 11, the packet information obtaining unit 12 records the receiving time and the packet data flow corresponding thereto; when the image frame packet is not When the first image frame received by the receiving unit 11 is encapsulated, the packet information obtaining unit 12 calculates the packet interval time in addition to the receiving time and the packet data flow corresponding thereto. The packet interval time is the difference between the receiving time and the previous receiving time.

As shown in step S3, the packet information obtaining unit 12 compares the packet interval time with the average picture interval time.

As shown in step S4, if the result of the comparison in step S3 is that the packet interval time is less than the average picture interval time, the packet information obtaining unit 12 obtains the accumulated data flow according to the packet data flow, and returns to step S1. The packet information obtaining unit 12 uses the following formula to obtain the accumulated data flow:

S _ curr = S _ curr + S _ p ;

Where S _ curr represents the accumulated data traffic and S _ p represents the packet data traffic.

As shown in step S5, if the result of the comparison in step S3 is that the packet interval time is greater than or equal to the average picture interval time, the packet information obtaining unit 12 uses the accumulated data traffic at this time as a picture data flow of an image frame.

That is, when the packet interval time is greater than the average picture interval time, theoretically, the two image frame packets received in sequence corresponding to the packet interval time should belong to different image frames, so the packet information obtaining unit 12 will The accumulated data flow at this time is taken as the screen data flow of the image screen. Taking a picture packet of a video stream (Video Stream) 3 as an example (see FIG. 3), when the interval time of the packets of the two image frame packets 31 and 32 is greater than the average picture interval time, it represents the image frames. The packets 31, 32 belong to different image frames.

As shown in step S6, the packet information obtaining unit 12 updates the average picture interval time according to the average picture interval time, the packet interval time, a third weight value, and a fourth weight value. The packet information obtaining unit 12 updates the average picture interval time by using the following formula: T _ avg = β × T _ avg + (1 - β ) × T _ curr ; where T _ avg represents the average picture interval time, T _ curr represents the packet interval time, β represents the third weight value, and (1- β ) represents the fourth weight value. In the preferred embodiment, the more β is closer to 1, the more the long-term average condition can be simulated; conversely, if β is closer to 0, the current actual situation can be reflected.

As shown in step S7, the packet information obtaining unit 12 updates the average data traffic according to the average data traffic, the accumulated data traffic, a first weight value, and a second weight value. The packet information obtaining unit 12 updates the average data flow by using the following formula: S _ avg = α × S _ avg + (1 - α ) × S _ curr ; where S _ avg represents the average data flow, α represents The first weight value, and (1- α ) represent the second weight value. Similarly, the closer α is to 1, the more it can simulate the long-term average; on the other hand, if α is closer to 0, the more the current situation is reflected.

As shown in step S8, the packet information obtaining unit 12 adds 1 to the count value.

As shown in step S9, the packet information obtaining unit 12 determines whether the count value is greater than a certain number of times N; if so, proceeds to step S10; if not, returns to step S1.

Finally, as shown in step S10, the picture type analyzing unit 13 determines a picture type corresponding to the image picture according to the picture data flow rate and the average data flow rate. That is, if the screen data traffic is greater than or equal to an average data traffic, step S11 is performed, that is, the screen type analyzing unit 13 determines that the screen type corresponding to the image screen is an I screen; if the screen data traffic is less than one The average data flow proceeds to step S12, that is, the screen type analyzing unit 13 determines that the screen type corresponding to the image screen is a P picture or a B picture. It is worth mentioning that, in step S12, the picture type analyzing unit 13 determines, according to a preset compression format (for example, H.261, MPEG4, etc.), that the picture type corresponding to the image picture belongs to This P picture is also the B picture.

In the preferred embodiment, the picture type analyzing unit 13 determines the picture type corresponding to each of the picture data traffic and the corresponding compression format. The details of the picture types corresponding to different compression formats are not the focus of the present invention, and can be easily understood by those skilled in the art, and thus are not described herein.

For example, the image packet of the video stream 3 is continued as shown in FIG. 3, and includes the image frame packet 31 corresponding to the same image frame, and the corresponding picture type is an I picture; The picture packet 32 corresponds to the video picture packets belonging to the same video picture, and the corresponding picture type is a B picture.

In summary, by determining the size relationship between the data flow of the picture and the average data flow, the picture type of the corresponding image frame (ie, video information) can be obtained; and the application related to the video processing The program does not need to fill in the relevant information in the instant protocol header according to the image standard of the image picture; therefore, the network endpoints in the transmission process do not need to have the ability to parse the network packet with the instant protocol header or the video information. Therefore, the object of the present invention can be achieved.

The above is only the preferred embodiment of the present invention, and the scope of the invention is not limited thereto, that is, the simple equivalent changes and modifications made by the scope of the invention and the description of the invention are All remain within the scope of the invention patent.

1. . . Video information acquisition system based on video data traffic

11. . . Receiving unit

12. . . Packet information obtaining unit

13. . . Picture type analysis unit

S0~S12. . . step

3. . . Video streaming

31. . . Image frame packet

32. . . Image frame packet

1 is a system diagram illustrating a preferred embodiment of a video information obtaining system based on video data traffic of the present invention;

2 is a flow chart illustrating steps of a method for obtaining video information based on video data traffic of the present invention; and

Figure 3 is a schematic diagram showing an example of a video stream in the preferred embodiment.

S0~S12. . . step

Claims (10)

A method for obtaining video information based on video data traffic is performed by using a computer, and the method comprises the following steps: (A) receiving an image picture packet, and recording a receiving time and a packet data flow corresponding to the image picture packet; B) calculating, according to the receiving time, a packet interval time corresponding to a previous receiving time of the previous image frame packet corresponding to the image frame packet; (C) comparing the packet interval time with an average picture interval time; If the packet interval time is less than or equal to the average picture interval time, obtaining a cumulative data flow according to the packet data flow, and returning to step (A); (E) if the packet interval time is greater than the average picture interval time, Then, the accumulated data flow at this time is used as a picture data flow of an image frame; and (F) determining a corresponding picture type of the image picture according to the flow of the picture data. According to the method for obtaining video information based on video data traffic according to claim 1, wherein in step (F), if the screen data traffic is greater than an average data traffic, determining that the image mode corresponding to the image image is An I picture, otherwise a P picture or a B picture. According to the method for obtaining video information based on video data traffic according to claim 1, the step (E) and (F) further includes a step (G): according to the average data flow, the accumulated data flow, and The first weight value, and a second weight value update the average data flow. The method for obtaining video information based on video data traffic according to item 3 of the patent application scope, wherein in step (G), the average data flow is updated by using the following formula: S_avg = α × S_avg + (1-α) × S_curr ; where S_avg represents the average data flow, S_curr represents the accumulated data flow, α represents the first weight value, and (1-α) represents the second weight value. The method for obtaining video information based on video data traffic according to item 4 of the patent application scope, wherein in step (D), the accumulated data flow is obtained by using the following formula: S_curr = S_curr + S_p ; wherein S_p represents the Packet data flow. According to the method for obtaining video information based on video data traffic according to item 3 of the patent application scope, the following steps are further included between steps (E) and (F): (I) according to the average picture interval time, the packet interval time, A third weight value and a fourth weight value update the average picture interval time. The method for obtaining video information based on video data traffic according to item 6 of the patent application scope, wherein in step (I), the average picture interval is updated by using the following formula: T _ avg = β × T _ avg + (1-β)× T _ curr ; where T _ avg represents the average picture interval time, T _ curr represents the packet interval time, β represents the third weight value, and (1-β) represents the fourth weight value . According to the method for obtaining video information based on video data traffic described in claim 6 of the patent application, steps (G) and (I) further include a step (J): repeating steps (A) to (I) to perform a specific frequency. The method for obtaining video information based on video data traffic according to item 8 of the patent application scope, wherein in step (F), judging according to a plurality of picture data flows and a preset compression format, determining a plurality of video frames respectively Corresponding majority of screen types. A video information obtaining system based on video data traffic includes: a receiving unit for receiving a plurality of image frame packets; and a packet information obtaining unit for calculating a picture data flow according to the image frame packets, wherein For the image frame packet, the packet information obtaining unit is configured to: record a receiving time corresponding to the image frame packet and a packet data flow; and respectively package the image frame corresponding to the image frame and the adjacent image frame. Waiting for the receiving time, calculating a packet interval time; obtaining a cumulative data flow according to the packet data flow; determining whether to use the accumulated data traffic as a picture data flow of an image frame according to the packet interval time; and a picture type The analyzing unit is configured to determine a picture type of the image frame according to the data flow of the picture.