Research of procedures for disintegration of the series length array as а mechanism way to confront unauthorized extraction of steganocontent

Размещено на http://www.allbest.ru/

V.N. Karazin Kharkiv National University

Research of procedures for disintegration of the series length array as а mechanism way to confront unauthorized extraction of steganocontent

Honcharov Mykyta Oleksandrovych Graduate student of the Faculty of Computer Sciences (specialty 125 - Cybersecurity),

Malakhov Serhii Vitalyovich Candidate of Technical Science, Senior researcher, Associate Professor of the Faculty of Computer Sciences

Kharkiv

Abstract

The paper presents the results of modeling attempts to unauthorized extract steganocontent (halftone test images) for different variants of disintegration of an array of the lengths series base block (BB) of the source images. The software model of the experimental algorithm provides for the possibility of changing the parameters of the spatial positioning of content BB. As part of the modeling, it is assumed that the attacker has successfully determined one of the two content processing parameters: - the current scanning scheme series BB and/or the current variants of the spatial orientation BB. As an example, one of the variants of the spatial positioning of BB is presented. Manipulations with the parameter of the spatial orientation of BB strengthen the opportunities to counteract unauthorized data extraction attempts. The modeling was performed on examples of two types of images: - portrait and landscape. The results of the attack of content are analyzed for various options for the disintegration of the source array of the lengths series BB was performed. Samples of attacked test images are presented. Attention is drawn to the fact that the use of the parameter «the spatial processing of BB» is an effective and computationally «simple» way to counteract attempts at unauthorized data extraction. It is concluded that entering into the structure key of the data extractor, the parameter «the spatial processing of BB», strengthens the resistance of content to attacks. The used methods of disintegration of the source array of the lengths series BB determine the structure of visual artifacts. Observed distortions (in case of unsuccessful selection of processing parameters) do not allow identification of content at the level of image type classification. The obtained results confirm that the use of disintegration procedures of the source array of series BB additionally complicates the work of the steganoanalyst and determines the further course of the attack.

Keywords: content, steganography, attack, run-lengths encoding; image; scan; spatial orientation; encoding with conversion; encapsulation; base block; data extraction.

Анотація

Гончаров Микита Олександрович аспірант факультету комп'ютерних наук (спеціальність 125 - Кібербезпека), Харківський національний

університет імені В.Н. Каразіна, м. Харків

Малахов Сергій Віталійович к.т.н., ст. науковий співробітник, доцент факультету комп'ютерних наук, Харківський національний університет імені В.Н. Каразіна, м. Харків

ДОСЛІДЖЕННЯ ПРОЦЕДУР ДЕЗІНТЕГРАЦІЇ МАСИВУ ДОВЖИН СЕРІЙ ЯК МЕХАНІЗМУ ПРОТИДІЇ НЕАВТОРИЗОВАНОЇ ЕКСТРАКЦІЇ СТЕГАНОКОНТЕНТУ

В роботі представлені результати моделювання спроб несанкціонованого вилучення стеганоконтенту (напівтонових тестових зображень) для різних варіантів дезінтеграції масиву довжин серій опорних блоків (ОБ) вихідних зображень. Програмна модель дослідного алгоритму передбачає можливість зміни параметрів просторового позиціонування ОБ контенту. В рамках моделювання зроблено припущення, що атакуючий вдало визначив один із двох параметрів обробки контенту: - поточну схему розгортки серій ОБ та/чи діючий варіант просторової орієнтації ОБ. В якості прикладу представлений один із варіантів просторового позиціонування ОБ. Маніпуляції з параметром просторової орієнтації ОБ, посилюють можливості із протидії спробам неавторизованого вилучення даних. Моделювання проводилося на прикладах двох типів зображень: - портрет та пейзаж. Виконано аналіз результатів атаки контенту для різних варіантів дезінтеграції вихідного масиву довжин серій ОБ. Представлені зразки атакованих тестових зображень. Звернено увагу, що використання параметра «просторової обробки

ОБ» є ефективним та обчислювально «простим» способом з протидії спробам неавторизованого вилучення даних. Зроблено висновок, що введення до структури ключа екстрактору даних, параметру «просторової обробки ОБ» посилює стійкість контенту к атакам. Використовувані способи дезінтеграції вихідного масиву довжин серій ОБ, визначають структуру візуальних артефактів. Спостережувані спотворення (при невдалому доборі параметрів обробки) не дозволяють ідентифікувати контент на рівні класифікації типу зображень. Отримані результати підтверджують, що використання процедур дезінтеграції вихідного масиву серій ОБ додатково ускладнює роботу стеганоаналітика та визначає подальший хід атаки.

Ключові слова: контент, стеганографія, атака, кодування довжин серій; зображення; розгортка; просторова орієнтація; кодування з перетворенням; інкапсуляція, опорний блок; екстракція даних.

Problem statement

One of the promising areas of providing concealment of the facts of information transmission and storage is the use of various steganographic methods. The ability to use the properties of digital content allows us to more effectively solve the issues of covert transmission, storage, and counteraction to unauthorized extraction of target information (content). Regardless of the direction of steganography, it is necessary to ensure the following: 1 - minimization of unmasking anomalies in data storage media (containers); 2 - supporting a certain level of content resistance to attempts of unauthorized extraction and/or containers distortion.

When hiding useful information in digital images (primarily in images - containers), certain distortions may occur.

By balancing the settings of the steganographic insertion videodata algorithm (hereinafter referred to as the «algorithm»), it is possible to ensure the level of distortion of the used containers, at a level, which is located below the threshold of sensitivity of the human visual system. [1,2]. By this ensures a decrease in the level of demasking anomalies containers. Correct parameters of the algorithm provide: - complication of the «work» of the steganoanalyst; - a balance is required between saving the characteristic properties of containers and the amount of permissible distortion for a specific type of content [3,4].

Analysis of the recent research

The presented modeling results are part of a general cycle of research aimed at further improving the procedures and determining the necessary parameters of videodata processing when implementing the chosen concept of a hybrid steganoalgorithm [4].

The main specificity of the experimental algorithm is its positioning for use as part of mobile platforms. This necessitates the determination of appropriate conditions to ensure the required compromise of characteristics at each stage of processing. [4-5]. The main components of such a compromise are: - the complexity of data processing; - the resistance of the content to attempts to «hack» it is ensure; - the acceptable level of visually noticeable [1-2] distortions container; - an acceptable level of content deformation; - a convenient and simple interface with all the necessary functions.

The use of a phased approach to modeling allows for identifying and taking into account specific features and parameters of data processing. This ensures the harmonization and coordination of the entire sequence of procedures to implement the overall plan. [4,6]. As part of the research, modeling of various conditions of compromising of the main protection mechanisms was carried out. Various combinations act as such conditions: 1 - schemes scanning of series BB; 2 - the methods of the spatial positioning BB. The results of content extraction obtained in the framework of these studies are presented in [7,8].

The purpose of the article is to summarize the results, modeling attempts of unauthorized extraction steganocontent (i.e., its attacks) when using different options for disintegration of the lengths series BB source array of the image - content [4,7-8].

Summary of the main material

The paper presents the results of attempts to attack the content in the assumption that the attacker managed to determine the current parameters of content processing, which are implemented at two main levels of protection (inter-block and intra-block, Fig. 2-4 [5]). The most indicative (in terms of view of visibility consequences) parameters of data processing were used within the conducted modeling. This makes it easier to understand the observed processes and assess the nature of content distortion.

The simplified diagram in Fig. 1 explains the general essence of the main processing procedures at two levels of protection at once. Features of the implementation of data processing procedures at each level of multiplexing (interblock and intra-block) in the conditions of using stacks of sampling parameters of series (BB and their length of series for the 1st level and the parameter of average brightness of BB, at the 2nd level ofprotection) of different lengths are considered in works [9-12]. The paper [13] considered the characteristic results of attack content attempts when both levels of protection were compromised at once in the conditions of using different methods of scanning series BB [4].

In Fig. 1, the block «Multiplex Level 1» (step №4), corresponds to the interblock processing mode (this is the 1st level of protection), and the block «Multiplex Level 2» (step №6), the intra-block multiplexing mode current parameters of content (2nd level of protection).

The results of attempts to attack the test image (portrait) when implementing different methods of scanning series BB [10,13] when applying the procedures of the inter-block security level (step 4 in Fig. 1) for sample stacks of different lengths [9] are shown in Fig. 2 (where the 1st column is a short stack, the 2nd column is a long stack). The results of attempts to attack the test image when different methods of The results of an image attack in the mode of two-level multiplex content when different methods of scanning series and stacks of multiplexing parameters (BB and their length of series for the 1st level and the average brightness parameter of BB at the 2nd level of protection) are shown in Fig. 4.

The results of a test image attack when both levels of protection are simultaneously compromised for different methods of scanning series and stacks of sampling current multiplexing parameters are considered in [13]. Thus, a situation was modeled in which the attacker made a mistake only in the method series BB was scanned. According to the modeling results, it was concluded that regardless of the options for processing content at any of the two levels of protection, an error in the current method of scanning (sampling) series BB does not allow the attacker to identify the content (see Fig. 1 in [13]).

Fig.1. Basic procedures when modeling attempts unauthorized extraction content

Figs. 2-4 reflect the results of the unsuccessful selection of the current processing parameters only at two of the provided levels of protection [4] when various attack scenarios.

Fig. 2. Inter-block level ofprotection for different methods of scanning series BB

(Pz = 7 [4]; The dimension of BB is 8*8 el.)

It should be emphasized that the methods of scanning and multiplexing the current parameters of the series (at two levels of protection) were used according to a simplified scheme (see Fig. 1 and Fig. 6 in [9]), i.e., as a demonstrator of the effects obtained. In general, the use of the principle of encoding series lengths [1] in the implementation of inter-block multiplexing procedures (step №4, Fig. 1) creates a good basis for countering attempts to unauthorized remove content (see Fig. 2), and has low computational complexity, which is important for adapting the algorithm to the conditions of mobile platforms.

Fig. 3. Intra-block level ofprotection for different methods of scanning series BB

(Pz = 7, The dimension of BB is 8*8 el.)

Manifestations of the symmetry of the artifacts observed in Fig. 2 (doubling, horizontal and vertical displacement of individual fragments) are caused exclusively by a simplified scheme for sampling the current parameters of series BB. The term «simplified» should be understood as a small period and an unchanged mask of mixing and shifting [9-10].

Fig. 4. Two-level multiplex ofprotection for different ways of scanning series BB

(Pz = 7, The dimension of BB is 8*8 el.)

From the obtained results, the effect of increasing the fragmentation of the source content is clearly observed. The process increases when the complexity of the multiplexing mask used increases, which confirms the importance of multi-stage processing, where each level of protection significantly increases the overall resistance to content hacking. This process is well followed by the complexity of the structure of the artifacts of the attacked samples (Fig. 2-4).

Below are the results of modeling attempts to unauthorized extraction of content under the condition of changing the spatial positioning of existing BB [6], and precedes the main procedures of the research algorithm (steps №№4-6 in Fig. 1). An assumption is made about the simultaneous compromise of the two main levels of protection (inter-block and intra-block multiplexing of the current parameters of series BB, according to steps №4 and №6 in Fig. 1).

Thus, the primary attention is on modeling possible attacks in relation to the two components of the data extractor key (Fig. 1 in [14]):

an element key that defines the current BB scan scheme [7];

an element key that determines the active realize of the spatial processing BB. disintegration halftone test image

The essence of the used manipulations with the spatial orientation parameter of BB is shown in Fig. 5. Thus, in accordance with the idea, when encoding content, all odd blocks are rotated to the left by 90° (marked as Rotate 90 °, and all even blocks are horizontally mirrored (Mirror). Thus, after the formation of the array of series BB, using the appropriate scanning scheme (marked as «The source array of series BB» in Fig. 5), the initial orientation of BB is changed (marked in Fig. 5 as «Array of series BB after...»). At the same time, the simplest test encoding scheme was used, which is repeated without changes for each subsequent pair of BB.

Fig. 5. Examples of the spatial processing BB

Taking into account possible attack combinations, the general modeling scheme is as follows:

- successful scan selection, but an error in the spatial orientation of BB;

- successful spatial orientation of BB, but an error in the scan.

The corresponding attack results are shown in Fig. 6(c-d). These samples correspond to the following parameters of the source content preprocessing:

the dimension of the blocks of the smoothing matrix is 3*3 el;

3rd preprocessing option (i.e., all with all) [4];

threshold value of the difference in brightness of elements of neighboring blocks of the original image, Pz=3 [15].

The «Pz» parameter remained the same both at the stage of smoothing the original image (step 2 in Fig. 1) and at the stage of forming series BB according to step 1 in Fig. 1. In both cases, the same scan scheme was used - a «Double Zigzag» («Double Zigzag» in Fig. 1 in [16]).In other words, in general, all the main tuning parameters of the algorithm were in the «comfortable» (in terms of visual artifacts and computational complexity) range of values.

Fig. 6. Results of content recovery under various attack combinations

As follows from the results, in both attack scenarios, the non-compromised part of the extractor key (highlighted in blue letters in Fig. 5) makes further identification of the objects of the attacked content impossible.

A pairwise comparison of the attacked samples ((a) and (b) in Fig. 6) follows that the structure distortion and the intensity of fragmentation of the attacked images (samples (c), (d) in Fig. 6), deprive the attacker of the opportunity to classify the type of content removed, in both cases of attack.

Fig. 7 shows a visualization of the existing difference between the original and attacked images for the indicated above smoothing parameters [4], but in the conditions of simultaneous error; - the current scan parameters; - the spatial orientation of BB (i.e., false rotation and reflection of BB (see Fig. 5)).

Fig. 7. The structure of content distortion with a simultaneous error in determining the scan and orientation of BB (BB 12*12 el., Pz=5)

In this case, the brighter the point or any image fragment (samples (b) and (d) in Fig. 7), the greater the difference between the attacked content and its original. Characteristic examples of an unsuccessful selection of the current parameters of the series scanning without any manipulations with the spatial orientation of BB, with a simultaneous compromise of two levels of protection at once (inter-block and intrablock), are presented in works [14,16-17]. Comparison of the images in Fig. 7(b,d) with their originals demonstrates that the presence of a large number of dark elements in the attacked content does not contribute to the successful identification (although the obtained PSNR values do not exclude this possibility [7]).

In Fig. 7(b), special attention should be paid to the area of the image, which is highlighted with a yellow marker. This area (finely dispersed granularity) differs from the rest, the peripheral part of the image. It is noticeable that in this case, the structure of the «grain» of errors is proportional to the size of the used blocks. This confirms the presence of a large number of BB with single or very short lengths of the formed of series BB [6]. This image fragment coincides with the face area, i.e., the most informative part of the test image [1]. In other words, within the «yellow area» in Fig. 7(b), there are mostly blocks whose content is very close to their original, which cannot be said by the results of a visual assessment of the attacked samples (in Fig. 7(a) and (c), marked with a red frame).

Thus, the used processing algorithm ensures the lowest level of distortions in the structure of BB, which forms highly detailed fragments of images. At the same time, the protection of content in these areas is ensured by the implementation of appropriate scanning schemes and changes in the spatial orientation of BB. The attacker's mistake in the two specified parameters only increases the overall effect (see the samples marked with a red frame in Fig. 7(a) and 7(c)).

A similar situation is observed for the test image of the «landscape» type in Fig. 7(d). In this example, a yellow marker outlines the fragments of the image with the most noticeable distortions, which correspond to the least informative areas of the content: 1 - the area of the sky above the city; 2 - the area of the water surface (with the least pronounced light-shadow boundary and the reflection of urban buildings on the water). That is, in these fragments of the image, the «work» of the algorithm for the formation «long» series BB is most noticeable. Conversely, in the rest of the image, characterized by a large number of small details, «fine grain» is again observed. Comparing the location of both «yellow fragments» in Fig. 1(d) with the original composition of the corresponding test image, the general relationship and distribution of characteristic distortions over the entire field of the attacked image are clearly visible. That is, the data of the «yellow» area of the image contain a large number of short length series, and the main contribution to the process of content fragmentation is provided by the scanning and the spatial positioning of BB.

Conclusion

The «series length» parameter of BB significantly destroys the correlations of the elements of the source array series of the image - content, which does not allow identifying the attacked samples of the original images.

Complications of the procedure for mixing the current encoding parameters significantly complicate the possibility of unauthorized content extraction (see comparison of different image columns in Fig. 2-4).

The conducted modeling is of a demonstration nature and aims to confirm the main assumptions regarding the selected data processing modes in the conditions of combining different schemes of series scanning and the spatial orientation of BB.

Taking into account the low computational complexity of BB scan and the potential variety of options for its implementation [10], it should be emphasized that the corresponding element in the structure key of the data extractor [4] in its significance, is not inferior to both procedural levels of processing [13].

The conducted modeling confirms that changes in the spatial orientation of the formed BB and the use of more complex scanning methods are effective tools for strengthening the content's resistance to unauthorized extraction attempts.

1. Adding an element to the data extractor key structure [4], which defines the spatial processing parameters of BB, allows us to reduce the requirements for the complexity of the used scan schemes.

8. The used processing modes ensure a low level of distortion in the blocks that form highly detailed image fragments. Protection content in such areas is ensured by using appropriate scanning schemes and changing the spatial orientation of BB. An attacker's mistake in the two specified parameters immediately increases the overall destructive effect (content fragmentation).

References

1. Pratt U. (1985). Tsyfrovaia obrabotka yzobrazhenyi [Digital Image Processing]. Moscow: Myr, Vol.1-2, pp. 136 [in Russian].

2. Zubarev Yu. B., Dvorkovych V. P. (1997). Tsyfrovaia obrabotka televyzyonnbikh y kompiuternbikh yzobrazhenyi [Digital processing of television and computer images]. M: MTsNTY, pp. 212 [in Russian].

3. Hrybunyn, V.H., Okov, Y.N., & Turyntsev, Y.V. (2002). Tsyfrovaia stehanohrafyia [Digital steganography]. Moscow: Solon-Press, pp. 212 [in Russian].

4. Lesnaia Yu., Honcharov N., & Malakhov S. (2021). Otrabotka kontsepta mnohourovnevoho multypleksa dannbikh hybrydnoho stehanoalhorytma [Practicing of the concept of the multilevel multiplex data of the hybrid steganoalgorithm]. Materialy I Mizhnarodnoi naukovo-teoretychnoi konferentsii - Proceedings of the I International Scientific and Theoretical Conference «Current issues of science, prospects and challenges» (Sydney, 17 Dec. 2021), Australia: Collection of scientific papers «SCIENTIA». Vol.2. pp. 48-56.

5. Liesnaia Yu., Honcharov M., & Malakhov S. (2023). Rezultaty modeliuvannia sprob nesanktsionovanoho vyluchennia stehanokontentu dlia riznykh kombinatsii ataky doslidnoho stehoalhorytmu [The results modeling attempts of unauthorized extraction steganocontent for various combinations of the attack of the research steganalgorithm]. Materialy II Mizhnarodnoi naukovo- praktychnoi konferentsii - Proceedings of the II International Scientific and Practical Conference «Science in the Environment of Rapid Changes» (Brussels, 6-8 Feb. 2023), Belgium: Scientific Collection «InterConf», (141), pp. 338-345.

6. Honcharov N., Lesnaia Yu., & Malakhov S. (2022). Adaptatsyia pryntsypa kodyrovanyia dlyn seryi dlia protyvodeistvyia popbitkam neavtoryzovannoi эkstraktsyy stehanokontenta [Adapting principle of series length encoding to counteract attempts unauthorized extraction steganocontent]. Grail of Science, (17), pp. 241-247.

7. Honcharov M.O., & Malakhov S.V. (2023). Doslidzhennia sposobiv rozghortky vykhidnykh blokiv zobrazhennia-stehanokontentu yak mekhanizmu protydii vid nesanktsionovanoi ekstraktsii danykh [Research the methods scanning of output blocks image-steganocontent as a protection mechanism against unauthorized data extraction]. Science and technology today, №4 (18), pp. 293-308.

8. Honcharov M.O., & Malakhov S.V. (2023). Modeliuvannia sprob nesanktsionovanoho vyluchennia stehanokontentu pry zmini parametriv prostorovoi oriientatsii opornykh blokiv kontentu [Modeling attempts to unauthorized extraction steganocontent when changing the spatial orientation parameters of the content base blocks]. Science and technology today, №6(20), pp. 114-129.

9. Honcharov M., Lesnaia Yu. (2022). Vykorystannia parametriv dovzhyn serii, yak elementa mizhblochnoho multypleksu danykh stehanoalhorytmu [Using the parameters of the series lengths as an element of the interblock data multiplex of the steganoalgorithm]. Computer Science and Cyber security, (1), pp. 30-38.

10. Honcharov N.A., Lesnaia Yu.E., Semenov A.S., & Malakhov S.V. (2022). Rezultatbi ataky stehanokontenta pry raznoi shyryne steka vbiborky seryi na эtape mezhblochnoho multypleksa dannbikh [The results of the attack of the steganocontent at different stack widths of series sampling at the stage of interblock multiplex data]. Materialy I Mizhnarodnoi naukovo-praktychnoi konferentsii - Proceedings of the I International Scientific and Practical Conference «Current trends in the development of modern scientific thought» (Haifa, 27-30 Sep. 2022), Israel: International Science Group, pp. 465-471.

11. Honcharov N., Lesnaia Yu, & Malakhov S. (2022). Modeliuvannia korotkoho steku vybirky dovzhyn serii pry riznykh kombinatsiiakh parametriv multypleksuvannia [Simulation of a short stack of sampled burst lengths with different combinations of multiplexing parameters]. Materialy VI Mizhnarodnoi naukovo-praktychnoi konferentsii - Proceedings of the VI International Scientific and Practical Conference «Multidisciplinary scientific notes. Theory, history and practice» (Edmonton, 1-4 Nov. 2022), Canada: International Science Group, pp. 684-687.

12. Lesnaia Yu.I, Honcharov M., & Malakhov S. (2022). Rezultatbi vnutryblochnoho multypleksyrovanyia parametra srednei yarkosty opornbikh blokov stehanokontenta naraznoi baze perestanovok [The results interblock multiplexing of the parameter average brightness base blocks steganocontent on different the permutation base]. Materialy IV Mizhnarodnoi naukovo- praktychnoi konferentsii - Proceedings of the IV International Scientific and Practical Conference «Debats scientifiques et orientations prospectives du developpement scientifique» (Paris, 11 Nov. 2022), France: Collection of scientific papers «АОГОЕ», pp. 78-81.

13. Lesnaia Yu., Honcharov N., & Malakhov S. (2022). Sposobbi razvertky parametrov seryi opornbikh blokov yzobrazhenyi, kak эlement sostavnoho kliucha эkstraktora dannbkh stehoalhorytma [The Methods scanning of the series parameters of image base block. as an element of the composite key of the stegoalgorithm data extractor]. Grail of Science, (23), pp. 254-258.

14. Liesnaia Yu., Honcharov M., Azarov S., & Malakhov S. (2023). Vizualizatsiia sprob nesanktsionovanoi ekstraktsii stehanokontentu pry pomylkovomu vyznachenni diiuchykh sposobiv rozghortky serii [Visualization attempts at unauthorized extraction of the steganocontent at erroneous determination the active methods of scanning series]. Grail of Science, (24), pp. 335-340.

15. Honcharov N., Lesnaia Yu, & Malakhov S. (2022). Modelyrovanye ataky stehanokontenta pry hrubbkh otsenkakh podobyia yskhodnbikh dannbkh y raznoi baze steka vbborky seryi [Modeling steganocontent attack under coarse source data similarity estimates and different series sampling stack base]. Materialy III Mizhnarodnoi naukovo-praktychnoi konferentsii - Proceedings of the III International Scientific and Practical Conference «Scientific practice: Modern and classical research methods» (Boston, 16 Sep. 2023), USA: Collection of scientific papers «АОГОЕ», pp. 86-90.

16. Lesnaia Yu., Honcharov M., Semenov A., & Malakhov S. (2023). Modeliuvannia rozghortky serii opornykh blokiv zobrazhennia, yak instrumentu z protydii sprobam nesanktsionovanoi ekstraktsii stehanokontentu [Modeling the scanning series of base blocks an image as a tool for countering attempts at unauthorized extraction of the steganocontent]. Materialy IV Mizhnarodnoi naukovo-praktychnoi konferentsii - Proceedings of the IV International Scientific and Practical Conference «Grundlagen der modernen wissenschaftlichen forschung» (Zurich, 31 Mar. 2023), Switzerland: Collection of scientific papers «АОГОЕ», pp. 109-116.

17. Lesnaia Yu., Honcharov M., Malakhov S., & Melkozorova O. (2023). Rezultaty nesanktsionovanoi ekstraktsii stehanokontentu pry realizatsii dvokhprokhidnoi rozghortky serii vykhidnykh blokiv [The results unauthorized extraction the steganocontent when the implementing a two-pass scan series of output blocks]. Materialy III Mizhnarodnoi naukovo- praktychnoi konferentsii - Proceedings of the III International Scientific and Practical Conference «Ricerche scientifiche e metodi della loro realizzazione: esperienza mondiale e realta domestiche» (Bologna, 3 Mar. 2023), Italy: Collection of scientific papers «АОГОЕ», pp. 65-67.

Література

1. Прэтт У. Цифровая обработка изображений: монографія / Прэтт У.; пер. с англ. Д. С. Лебедева. - М.: Мир, 1985. -Т. 1,2. - 736 с.

2. Цифровая обработка телевизионных и компьютерных изображений : монографія / Зубарев Ю. Б., Дворкович В. П. - М.: МЦНТИ, 1997. - 212 с.

3. Цифровая стеганография : монографія / Грибунин В. Г., Оков И. Н., Туринцев И. В. - М.: Солон-Пресс, 2002. - 272 с.

4. Лесная Ю., Гончаров Н., Малахов С. Отработка концепта многоуровневого мультиплекса данных гибридного стеганоалгоритма. Матеріали І Міжнародної науково- теоретичної конференції. «Current issues of science, prospects and challenges» (м. Сідней, 17 груд. 2021 р.), Австралія: Збірник наукових праць «SCIENTIA», 2021. - Т.2. - C. 48-56.

5. Лєсная Ю., Гончаров М., Малахов С. Результати моделювання спроб несанкціонованого вилучення стеганоконтенту для різних комбінацій атаки дослідного стегоалгоритму. Матеріали ІІ Міжнародної науково-практичної конференції. «Science in the Environment of Rapid Changes» (м. Брюссель, 6-8 лют. 2023 р.), Бельгія: Науковий збірник «InterConf», 2023. - С. 338-345.

6. Гончаров Н. Адаптация принципа кодирования длин серий для противодействия попыткам неавторизованной экстракции стеганоконтента / Гончаров Н., Лесная Ю., Малахов С. // Grail of Science. - 2022. - №17. - С. 241-247.

7. Гончаров М.О. Дослідження способів розгортки вихідних блоків зображення- стеганоконтенту як механізму протидії від несанкціонованої екстракції даних / Гончаров М.О., Малахов С.В. // Наука і техніка сьогодні. - 2023. - №4(18). - С. 293-308.

8. Гончаров М.О. Моделювання спроб несанкціонованого вилучення стеганоконтенту при зміні параметрів просторової орієнтації опорних блоків контенту / Гончаров М.О., Малахов С.В. // Наука і техніка сьогодні. - 2023. - №6(20). - С. 114-129.

9. Гончаров М. Використання параметрів довжин серій, як елемента міжблочного мультиплексу даних стеганоалгоритму / Гончаров М., Лєсная Ю.. // Комп'ютерні науки та кібербезпека. - 2022. - № 1. - С. 30-38.

10. Гончаров Н.А., Лесная Ю.Е., Семенов А.С., Малахов С.В. Результаты атаки стеганоконтента при разной ширине стека выборки серий на этапе межблочного мультиплекса данных. Матеріали I Міжнародної науково-практичної конференції «Current trends in the development of modern scientific thought» (м. Хайфа, 27-30 вер. 2022 р.), Ізраїль: International Science Group, 2022. - С. 465-471.

11. Гончаров М., Лесная Ю., Малахов С. Моделювання короткого стеку вибірки довжин серій при різних комбінаціях параметрів мультиплексування. Матеріали VI Міжнародної науково-практичної конференції «Multidisciplinary scientific notes. Theory, history and practice» (м. Едмонтон, 1-4 лист. 2022 р.), Канада: Міжнародна наукова група (isg-konf.com), 2022. - С. 684-687.

12. Лесная Ю., Гончаров М., Малахов С. Результаты внутриблочного мультиплексирования параметра средней яркости опорных блоков стеганоконтента наразной базе перестановок. Матеріали IV Міжнародної науково-практичної конференції «Debats scientifiques et orientations prospectives du developpement scientifique» (м. Париж, 11 лис. 2022 р.), Франція: Збірник наукових праць «АОГОЕ», 2022. - С. 78-81.

13. Лесная Ю. Способы развертки параметров серий опорных блоков изображений, как элемент составного ключа экстрактора данных стегоалгоритма / Лесная Ю., Гончаров Н, Малахов С. // Grail of Science. - 2022. - №23. - С. 254-258.

14. Лєсная Ю. Візуалізація спроб несанкціонованої екстракції стеганоконтенту при помилковому визначенні діючих способів розгортки серій / Лєсная Ю., Гончаров М., Азаров С., Малахов С. // Grail of Science. - 2023. - №24. - С. 335-340.

15. Гончаров Н., Лесная Ю., Малахов С. Моделирование атаки стеганоконтента при грубых оценках подобия исходных данных и разной базе стека выборки серий. Матеріали III Міжнародної науково-практичної конференції «Scientific practice: modern and classical research methods» (м. Бостон, 16 вер. 2022 р.), США: Збірник наукових праць «ЛОГОЕ», 2022. - С. 86-90.

16. Лесная Ю., Гончаров М., Семенов А., Малахов С. Моделювання розгортки серій опорних блоків зображення, як інструменту з протидії спробам несанкціонованої екстракції стеганоконтенту. Матеріали IV Міжнародної науково-практичної конференції «Grundlagen der modernen wissenschaftlichen forschung» (м. Цюріх, 31 бер. 2023 р.), Швейцарія: Збірник наукових праць «ЛОГОЕ», 2023. - С. 109-116.

17. Лесная Ю., Гончаров М., Малахов С., Мелкозьорова О. Результати несанкціонованої екстракції стеганоконтенту при реалізації двохпрохідної розгортки серій вихідних блоків. Матеріали III Міжнародної науково-практичної конференції «Ricerche scientifiche e metodi della loro realizzazione: esperienza mondiale e realta domestiche» (м. Болонья, 3 бер. 2023 р.), Італія: Збірник наукових праць «ЛОГОЕ», 2023. - С. 65-67.

Размещено на Allbest.ru