IIARD International Institute of Academic Research and Development

INTERNATIONAL JOURNAL OF COMPUTER SCIENCE AND MATHEMATICAL THEORY (IJCSMT )

E-ISSN 2545-5699
P-ISSN 2695-1924
VOL. 11 NO. 2 2025
DOI: 10.56201/ijcsmt.v11.no2.2025.pg59.72

---

Chaos-Based Image Encryption Using Refined Hash Function

Ejiiobe Precious Nyimeobari

---

Abstract

This study presents an enhanced chaos-based image encryption framework that integrates refined hash functions to address modern challenges in secure image transmission and cryptographic robustness. Conventional encryption techniques often struggle to defend against sophisticated cyberattacks and ensure data integrity, especially in critical applications such as medical imaging and IoT security. The proposed encryption system combines the inherent unpredictability of chaotic systems with advanced hashing mechanisms to develop a robust security framework. A chaos generator and mapping function produce high-quality chaotic signals, which are transformed into the cryptographic domain to enhance encryption strength. The refined hash function incorporates image-specific attributes, generating unique and highly sensitive hash values capable of detecting even the slightest data alterations. The encryption system is implemented in Python, leveraging its extensive cryptographic libraries for precision and efficiency. Performance evaluation is conducted using statistical metrics such as histogram analysis, adjacent pixel autocorrelation, and key sensitivity tests, which reveal significant improvements in randomness and robustness against differential attacks. Experimental results indicate an NPCR of 99.81%, UACI of 33.69%, and Shannon entropy of 7.998, demonstrating superior encryption quality compared to traditional methods. Additionally, the proposed system achieves faster execution times, making it suitable for real-time applications requiring high security and computational efficiency. This research contributes to cryptography by proposing a scalable and efficient encryption model that mitigates existing vulnerabilities while ensuring robust protection of sensitive data. The practical applications of this system extend to secure communications, healthcare data protection, and IoT security, where confidentiality and integrity are pa

keywords:

Chaos, Image Encryption, Refined Hash Function

---

References:

collectively underscore the growing importance of encryption methods in diverse technological
domains.
The exponential growth of digital data transmission, particularly in sensitive domains like
healthcare and IoT, has amplified the demand for robust encryption techniques. Traditional
encryption methods (e.g., AES, DES) often falter against sophisticated cyberattacks due to their
linearity and deterministic nature. Chaos-based encryption, leveraging the inherent
unpredictability and sensitivity of chaotic systems, has emerged as a promising alternative.
However, challenges persist in ensuring data integrity, resisting differential attacks, and
maintaining real-time efficiency. This study addresses these gaps by integrating a refined hash
function into a chaos-based encryption framework, enhancing both security and computational
efficiency.
This work aims to Design a chaos-based image encryption system using refined hash function,
evaluate the system’s security through statistical metrics like NPCR, UACI, Entropy and also to
Compare its performance with existing methods in terms of execution time and robustness. The
proposed system enhances data confidentiality and integrity for applications like medical imaging
and IoT security. By combining chaotic dynamics with a hash function sensitive to minute data
alterations, it offers a scalable solution resistant to differential and brute-force attacks.
This study is important for several reasons. First, chaos-based encryption offers superior security
by leveraging its extreme sensitivity to initial conditions and the inherently unpredictable nature
of chaotic systems. Second, in an era where protecting sensitive information is critical, the study’s
findings pave the way for more effective encryption techniques that safeguard data across diverse
applications—from communication systems to financial transactions. Finally, as cyber threats
continually evolve, the optimized chaos-based encryption method proposed here serves as a
powerful tool in reducing risks such as unauthorized access, data breaches, and other malicious
activities.
RESEARCH METHOD
The design methodology for a chaos-based encryption system involves a constructive research
approach to conceptualize, develop, and implement the encryption technique. It encompasses the
systematic identification, selection, processing, and analysis of information pertinent to the
research. This methodology serves as a scientific framework for addressing the problem at hand,
guiding researchers in their exploration, description, explanation, and prediction of phenomena
related to chaos-based encryption.
At its core, the research methodology for chaos-based encryption entails studying the methods
through which knowledge is acquired and applied in the creation of the encryption system. Its
primary objective is to outline a comprehensive work plan for conducting the research, ensuring
that each step is carefully designed and executed to achieve meaningful results
2.1 EXPERIMENTAL DATA ANALYSIS
The dissertation utilizes several analyses which includes, Intensity histogram analysis, Adjacent
pixel autocorrelation test, and key sensitivity tests, to assess the efficiency and robustness of
encryption algorithms. The intensity histogram of the original image in Figure 1.0 graphically
depicts the spread of pixel intensities. The x-axis represents the spectrum of pixel values, often
ranging from 0 to 255 in grayscale images. Lower values indicate darker tones, while higher values
indicate brighter tones. The y-axis represents the number or frequency of pixels for each intensity
level, indicating the prevalence of certain intensities in the image. Within the framework of an
intensity histogram with a y-axis range of 0 to 1400, it indicates that the image exhibits a wide-
ranging distribution of pixel intensities, encompassing both dark and brilliant values. The pixel
count distribution extends up to 1400 pixels for a specific intensity level. Examining these
histograms is advantageous for comprehending the global contrast and tonal attributes of an image,
facilitating image processing endeavors such as contrast modifications or discerning distinct
features based on intensity patterns.
Figure 1.0: Original Image
Table 1.0: Histogram analysis of the original image
Pixel Value
Red Channel Count
Green Channel Count
Blue Channel Count
0
2081
2380
2408
1
145
169
150
2
136
157
153
3
141
153
153
4
140
148
142
5
113
119
123
6
105
101
99
7
94
118
103
8
143
159
142
9
106
117
124
Figure 1.1: Intensity of the original image (histogram)
Figure 1.2: Encrypted Image
Table. 1.2: Histogram Analysis Table of the Encrypted Image
Pixel Value
Red Channel Count
Green Channel Count
Blue Channel Count
0
2877
1967
1757
1
558
641
423
2
637
664
376
3
708
762
415
4
649
733
376
5
707
840
422
6
737
801
337
7
716
843
416
8
659
856
326
9
670
743
302
Figure 1.3: Intensity of the encrypted image (histogram)
The histogram offers a visual depiction of the distribution of pixel values throughout the full
intensity spectrum, providing insights on the general brightness and contrast characteristics of the
encrypted image. Examining the histogram allows for the identification of patterns, peaks, or
fluctuations in pixel intensities, which assists in evaluating the visual attributes of the image and
potential qualities affected by the encryption procedure.
2.2: STATISTICAL AND SECURITY ANALYSIS
The robustness of the encryption scheme was evaluated using several widely recognized metrics,
including the Number of Pixels Change Rate (NPCR), Unified Average Changing Intensity
(UACI), Peak Signal-to-Noise-Ratio (PSNR), and Shannon Entropy. These metrics assess how
effectively the encryption disrupts the original image’s structure, how sensitive the encryption is
to small changes, and the overall unpredictability of the encrypted image. The encryption achieved
an NPCR of 99.814% and a UACI of 33.694%, indicating high resistance to differential attacks.
The Shannon Entropy value of 7.998 demonstrates that the cipher image is highly random,
minimizing the possibility of leaking any useful information. Additionally, the PSNR of 7.723
indicates the image distortion introduced by encryption is substantial, further enhancing security.
This can be seen in Table 4.8
Table 2.0: Statistical and Security Analysis Metrics
Metric
Value
Number of Pixels Change Rate (NPCR)
99.814%
Unified Average Changing Intensity (UACI)
33.694%
Peak Signal-to-Noise-Ratio (PSNR)
7.723
Shannon Entropy
7.998
3.0: PROPOSED IMPROVED CHAOS-BASED IMAGE ENCRYPTION SYSTEM
The system architecture in Figure 3.0 incorporates a comprehensive approach to encryption,
leveraging chaos theory and refined hash functions to ensure robust data security. At its core, the
architecture comprises a chaos generator for producing high-quality chaotic signals, a mapping
function to transform these signals into a suitable domain for cryptographic operations, and a key
generation mechanism to generate secure cryptographic keys. Encryption and decryption processes
are facilitated through chaos-enhanced operations and the application of the refined hash function,
providing an additional layer of security. User interaction is facilitated through a user-friendly
interface, while secure key management ensures the protection of encryption keys. Rigorous
security measures, including entropy evaluation and cryptographic tests, are integrated into the
architecture to uphold data confidentiality and integrity. Overall, the architecture is designed to
deliver high-level security, performance, and reliability in encrypting and

DOWNLOAD PDF

---

Back