Content introduction chapter development of a complex image processing method

The erosion operation looks like this:
The expansion filter generates white only if all the pixels in the block are white. Otherwise, black is generated. Thus, as a result of applying the expansion filter, the white areas of the image are narrowed.
Definition of the dilation operation (expansion):
The beginning of the development of an application in the C++ high-level language in the Borland C++ Builder environment was carried out with the creation of a program form, in which the necessary components for working with the application were added. Figure 9 shows a fragment of the application form with buttons.
The form consists of a section panel, in which there are such items as "File" and "Effects". In the "File" section there are buttons with pictures that allow you to open, save an image and close the application. In the "Effects" section, the choice of filters is implemented accordingly. Figure 10 shows a fragment of the form in the "Effects" section.
Algorithms for image processing methods selected using a graphical editor are presented below.
After that, a morphological method for filtering dilation and erosion was implemented for the resulting image.  Comparing the obtained results, in the developed application it is noticeable that the median filter is processed much more slowly than in the graphical editor, in other methods the processing speed does not differ. The results of image processing in the developed application and the graphics editor are similar, which indicates good accuracy in writing program codes in the application.
The problem of identifying the unit of articulation of a linked text is one of the debatable ones in modern linguistics. Until now, there is no unity in determining the essence of the minimum constitutional unit of the text in oral speech. Linguists name a rhythmic group, an intonational group, a syntagma, a pausal group, a separate phrase, an utterance, a phonoparagraph, a superphrasal unity as a possible division unit. Note that the choice of division units largely depends on the goals and objectives of the study, the type of text.
A feature of the sounding informational monologue texts (in the present study - the news programs of the BBC and Independent News) is their two-sidedness. The text simultaneously exists in two forms: a prepared written text and its oral implementation. In phonetic literature, it is noted that during the division of sounding speech, and this primarily applies to prepared readable texts, syntactic and intonation division occurs. The former is based on syntactic links between segments of the speech sequence, and the latter is based on the assessment and attitude of the speaker to the utterance, depending on the conditions of the context and on the situation of communication. In many respects, this determines the selection of the next hierarchy of segmentation units, which will correspond to both syntagmatic and intonational division. On the linear axis, these units can be represented as follows: syntagma - phrase - phonoparagraph - text. To ensure the unambiguity of the presentation in this paper, it seems necessary to define these concepts.
Following Yu. A. Dubovsky [1, p. 19] by a phono-paragraph we mean “a hierarchically-limiting semantic-prosodic unit of an oral text capable of adequately representing the model of the prosodic structure of a certain type of text as a whole”. The analysis of news programs showed that each informational message is a text with a certain structure, the unity of the theme, absolute lexical-syntactic and intonational completeness. At the same time, in the vast majority of cases, these texts consist of one phono paragraph. Thus, it can be stated that in the texts studied in this work, the phono paragraph and the text of the information message often coincide (this remark seems to be especially important in the further classification of syntagmas).
The structural unit of a phono paragraph is a phrase. The phrase is understood by us as a semantic unity, the integrity of which is created by means of intonation, as well as a certain syntactic structure [2, p. 558].
It is known that a phrase consists of one or more intonation groups, which can be defined as the minimum unit included in the phrase and representing a speech segment, limited on both sides by pauses or their physical correlate, having intonational unity and completeness. The disadvantage of this concept is the fact that the intonation group can only correlate with the sounding text, while the object of this study is the text in its two implementations: oral and written. In our opinion, this requirement corresponds to the term "syntagma", which is wider than the term "intonation group" and is used in the study of both sounding and written speech. Consequently, the syntagma was chosen as the minimum unit of study, within the boundaries of which intonational modifications are considered, which we define following the concept of L. V. Shcherba [3, p. 86]. According to L. V. Shcherba, a syntagma is the simplest intonation unit of speech, the ultimate basic syntactic unit that expresses a single semantic whole in the process of speech-thought and is phonetically united by the effort of the last verbal stress, i.e. special syntagmatic stress. Syntagma can consist of a group of words and one word.
The traditional division of syntagmas into final and non-final occurs on the basis of their location in the phrase. However, it is known that not every final syntagma is completely complete in terms of content, and this primarily applies to those cases where the object of study is such a complex formation as a text. That is why in this paper it is proposed to divide syntagmas from the standpoint of their semantic-syntactic completeness/incompleteness into non-final, relatively final and final syntagmas. Note that non-final syntagmas will correspond to syntagmas that begin a phrase or are inside it, relative to final syntagmas - syntagmas that complete statements (phrase), and final - syntagmas that complete a phono-paragraph or the entire informational message.
It should be emphasized that, since the object of study in this work is texts, and not individual syntagmas or utterances, and, given the stylistic features of the studied information texts, in this work, non-final syntagmas will be defined as incomplete, and relatively final and final - as complete speech units. (smaller and greater degree of completion).
The above classification of syntagmas is based primarily on the principles of syntactic articulation. However, any oral realization of the text is impossible without taking into account the intonational features of non-final, relatively final and final syntagmas. It is noted in the linguistic literature that, as a rule, non-final syntagmas are formed in a tone that carries the semantics of incompleteness (for example, a low descending tone), relatively final syntagmas are characterized by both intonationally complete and incomplete formation, and final syntagmas are always formed by one of their descending tones, carrying the semantics of completeness. However, a comprehensive experimental study of the relationship between intonation and the semantic-syntactic factor (type of syntactic structure, semantic relations within the text and the functioning of the utterance) in two versions of the modern English language (in the general English intonation norm and the southern regional intonation standard), made it possible to distinguish the following types of syntagmas characteristic of the texts of information messages: non-final intonation-incomplete, non-final intonation-complete, relatively final intonation-incomplete, relatively final intonation-complete, final intonation-complete and final intonation-incomplete. It should be emphasized that almost all of the selected types of syntagmas are found in both variants of the English language under study, with the exception of the final intonation-incomplete syntagma, which is presented only in the southern regional pronunciation standard. made it possible to single out the following types of syntagmas characteristic of the texts of informational messages: non-final intonation-incomplete, non-final intonation-incomplete, relatively final intonation-incomplete, relatively final intonation-complete, final intonation-complete and final intonation-incomplete. It should be emphasized that almost all of the selected types of syntagmas are found in both variants of the English language under study, with the exception of the final intonation-incomplete syntagma, which is presented only in the southern regional pronunciation standard. made it possible to single out the following types of syntagmas characteristic of the texts of informational messages: non-final intonation-incomplete, non-final intonation-incomplete, relatively final intonation-incomplete, relatively final intonation-complete, final intonation-complete and final intonation-incomplete. It should be emphasized that almost all of the selected types of syntagmas are found in both variants of the English language under study, with the exception of the final intonation-incomplete syntagma, which is presented only in the southern regional pronunciation standard. relatively final intonation-complete, final intonation-complete and final intonation-incomplete. 
It should be emphasized that almost all of the selected types of syntagmas are found in both variants of the English language under study, with the exception of the final intonation-incomplete syntagma, which is presented only in the southern regional pronunciation standard. relatively final intonation-complete, final intonation-complete and final intonation-incomplete. It should be emphasized that almost all of the selected types of syntagmas are found in both variants of the English language under study, with the exception of the final intonation-incomplete syntagma, which is presented only in the southern regional pronunciation standard.
The following examples present the types of syntagmas that are typical for texts of informational messages, and are not typical for the implementation of individual phrases or phono paragraphs.
The American balloonist Steve Forset has been picked up by a yacht after crashing into the sea five hundred miles off north-eastern coast of Australia during the thunderstorm. Rescuers sailed through the night to reach his life raft (Independent Radio).
The nuclear zone of the non-final syntagm "The American balloonist Steve Forset has been picked up by a yacht..." is pronounced with a high falling tone. It is known that the descending tone has the semantics of completeness and forms relatively independent speech segments. However, the speaker reading the prepared written text has the only opportunity to highlight the most informative fragments - the use of intonational means. It is the high descending tone in this case that contributes to the markedness of some elements against the background of others, i.e. highlighting important words within a syntagma or syntagmas in the entire message.
Four former East-German officials have been found guilty of ordering guards to shoot people as they tried to cross the Berlin Wall. The former Chief of the Communist Party and three generals were given suspended jail sentences for their role in the killings (BBC News).
In this example, the terminal zone relative to the final syntagm "... as they tried to cross the Berlin Wall ..." is formed in a low descending tone. At the same time, the above syntagma cannot be defined as completely completed, since the following sentence develops the idea and provides additional information about both the status of the convicts and the punishment. In this case, the choice of a low descending tone may be associated with the stylistic features of the texts under study: clarity, accuracy of presentation, neutral sound.
syntactic intonation segmentation text
Gross domestic product that's the value of all goods and services produced in Britain grew by 2.6% in the second quarter of the year. The figures are in line with city forecasts and are seen as good news for the Chancellor (Independent Radio).
In this informational message, the final syntagm “…and are seen as good news for the Chancellor…”, being absolutely complete, is made out by the speaker who owns the southern pronunciation standard, an incomplete descending tone, which is traditionally considered a tone that conveys incompleteness.
The study of the interaction of syntactic and intonational means made it possible to present a more detailed classification of the minimum semantic units (syntagmas) that segment the sounding informational text in various versions of the modern English language. Both variants (general English intonation norm and southern regional intonation standard) are characterized by non-final intonation-complete and intonation-incomplete, relatively final complete and relatively final incomplete, final intonation-complete syntagmas. The revealed final intonation-incomplete syntagma is typical only for the southern regional variant of the English language. In addition, it should be emphasized that the nomenclature and frequency of terminal tones involved in the design of the nuclear zones of syntagmas will depend on the type of pronunciation.