In prior art, several methods for the transmission of video images in a data communication network are known. A feature typical of these methods is that the image is compressed in some way before transmission to allow faster transmission. Among the prior-art compression methods are e.g. MPEG (Moving Pictures Experts Group), and H.261 and H.263 standar- dized by ITU (International Telecommunication Union, ITU). The prior-art compression methods are characte- rized in that, instead of transmitting complete fra- mes, only the differences between successive frames are transmitted. Moreover, predictive coding is used to reduce temporal redundancy, and modification coding to reduce spatial redundancy.

A problem with the prior-art methods is that they apply compression to entire frames and transmit all differences between successive frames. Consequent- ly, the amount of data to be transmitted remains too high for the needs of narrow-band data communication networks. For example, if an MPEG-compressed video image with a resolution of 388*288 is transmitted over a GSM network, the transmission of a single frame may take as long as 15 seconds.

The object of the present invention is to present a new method that eliminates the above- mentioned drawbacks.

A specific object of the invention is to pro- duce a method for processing a video image so that the image can be readily transmitted even in a narrow-band data communication network, such as a GSM network.

As for the features characteristic of the

present invention, reference is made to the claims.

In the method of the present invention for the transmission of video images over a data communi- cation network, the video image is encoded at the sen- ding end, transmitted, and decoded at the receiving end. The equipment at the sending end comprises a re- ference memory, in which the first frame of the video image is stored. Next, the first frame is compressed and transmitted. At the receiving end, the frame is stored in an image memory. The first frame stored in the reference memory and the next frame in sequence are divided into blocks. Mutually corresponding blocks of the frames are compared with each other to determi- ne a block difference. If the block difference exceeds a specified threshold, then the difference between the blocks is compressed and transmitted. In addition, the block is stored in the reference memory. At the recei- ving end, the difference is integrated into the image to be presented.

In an embodiment of the method, the frame is divided into blocks having the size of 8*8 pixels.

In an embodiment of the method, the block difference is determined by comparing some of the mu- tually corresponding pixels contained in blocks cor- responding to each other and calculating the number of pixels differing from each other.

In an embodiment of the method, the compari- son of pixels is performed using some of the bits in- dicating pixel value.

In an embodiment of the method, the compari- son of pixels is performed based on the luminance com- ponent of the pixels.

In an embodiment of the method, a predeter- mined part of each frame is transmitted and integrated into the image to be presented.

In an embodiment of the method, the compres- sion method used is the JPEG (Joint Photographic Ex-

perts Group, JPEG) method.

As compared with prior art, the present in- vention has the advantage that it allows a conside- rable reduction in the amount of data to be transmit- ted, as the frames are divided into blocks and the difference between blocks corresponding to each other is only transmitted if the block difference is suffi- ciently large. This makes it possible to transmit vi- deo images in a narrow-band network as well. The met- hod is particularly well suited for the transmission of video images obtained from a mainly stationary ca- mera, such as a monitoring camera.

In the following, the invention will be described in detail by the aid of an application example by referring to the attached drawing, in which Fig. 1 presents a block diagram of the method of the invention; Fig. 2 presents a hardware assembly implemen- ting the method of the invention; and Fig. 3 presents an example of data flow transmitted by the method of the invention.

Fig. 1 is a block diagram presenting an example of the method of the invention. The parameter Yp, which indicates the value of the Y co-ordinate, is set to zero. The first frame of the video image is grabbed and stored in the reference memory. The frame is compressed by the JPEG method and transmitted. At the receiving end, the frame is decompressed, stored in the image memory and presented. The next frame is then grabbed and compared block by block with the first frame. The frames are divided into blocks of 8*8 pixels. In blocks corresponding to each other, the lu- minance component of every second pixel is compared using five most significant bits. If the absolute dif- ference is larger than two, then the block difference value is increased by one. If a block difference value exceeding 14 is obtained, then the difference between

the blocks being compared is large enough to be transmitted. In this case, the difference between the blocks is calculated, compressed by the JPEG method and transmitted together with position data. In addi- tion, the block under processing is stored in the re- ference memory. Further, a predetermined part of each frame is transmitted. The size of this part is x- resolution * 8 pixels. The position of the predeter- mined part is shifted by eight pixels downwards bet- ween successive frames, in other words the value of the parameter Yp is increased by eight for each frame until it reaches the y-resolution value, whereupon the parameter Yp is reset to zero.

Fig. 2 presents an example of a hardware as- sembly implementing the method of the invention. Con- nected to a television camera, such as a monitoring camera 1, is a device 2 that grabs the frames for pro- cessing. The device 2 is further connected to a cent- ral processing unit 3, which takes care of image pro- cessing at the sending end. The central processing unit 3 communicates with a reference memory 4, which is used for the determination of block difference. The central processing unit 6 at the receiving end decodes the received data. Connected to the central processing unit 6 is an image memory 5, which is used to store the image to be presented. Moreover, the central pro- cessing unit 6 is connected to a display unit 7 used to present the image.

Fig. 3 presents an example of the data flow to be transmitted. The frame in the example comprises three blocks whose block differences have been found to be large enough to make it necessary to transmit the difference between successive blocks. These diffe- rences make up 192 bytes of data, which is compressed by the JPEG method to 5 bytes. In addition to the data contained in each block, the data transmitted compri- ses a step number delta, which indicates the position

of each block relative to the beginning of the frame and whose value is determined as follows: 0 -> begin- ning of frame, 1 -> removed by 8 pixels, 255 -> remo- ved by 8*255 pixels from the beginning.

The invention is not limited to the examples of its embodiments described above, but instead many variations are possible within the framework of the inventive idea defined by the claims.