Abstract— Neural image enhancement is an essential project in photo processing which aims at improving the first-class of an picture by using casting off noise, improving assessment and increasing sharpness. In this paper, we advise a novel Fuzzy Neural community (FNN) technique using clustering for picture enhancement. The proposed approach combines the benefits of each Fuzzy logic and Neural Networks to correctly enhance pictures. The FNN technique first uses clustering to divide an input photo into smaller regions based on their traits. Then, a Fuzzy common sense gadget is carried out to each location to decide the most efficient enhancement parameters. In the end, a Neural network is educated to map the enter photo to the enhanced version based at the decided parameters.one of the key features of the proposed approach is using clustering to phase the input photo. Conventional picture enhancement techniques observe an international enhancement to the whole picture without thinking about the spatial and contextual records present in one of kind regions. By means of dividing the image into smaller areas, the FNN method can perform enhancement on a nearby stage, resulting in extra accurate and optimized enhancements. Moreover, the clustering step reduces the computational complexity as it gets rid of the want for the FNN to technique the whole picture. The fuzzy common sense device used on this method provides a bendy approach to coping with imprecise and uncertain information present in picks.

Introduction

Photograph enhancement is the method of improving the satisfactory or visual appearance of a photo. [1].It may be finished through numerous strategies, together with filtering, side detection, and comparison enhancement. [2].But, traditional picture enhancement strategies won’t continually produce first-rate outcomes, especially on low-high-quality or blurry snap shots. [3].This has caused the improvement of extra advanced techniques, together with neural community-based picture enhancement. [4].One such approach that has proven promising consequences is the usage of Fuzzy Neural Networks (FNN) blended with clustering algorithms. [5].Fuzzy neural networks are neural networks that use fuzzy logic to handle uncertainty and imprecision, even as clustering algorithms are used to group similar information factors collectively. by using combining these two techniques, FNN-primarily based image enhancement techniques can correctly address the demanding situations posed by way of noisy and distorted pixie FNN-primarily based image enhancement, the network is trained to map the enter photo to a favored output picture. [6].This is performed by the usage of an aggregate of fuzzy common sense guidelines and neural network mastering algorithms. [7].the bushy common sense policies are used to address the uncertainty and imprecision in photo statistics, at the same time as the neural network getting to know algorithms are used to optimize the parameters of the community to acquire the desired output.[8].one of the key benefits of the usage of FNN for image enhancement is its potential to address non-linear relationships between the input and output snap shots are an important shape of speaking ideas, emotions, and messages.[9]. In state-of-the-art digitally-driven international, the first-class of photos has grown to be critical, whether or not it’s far for advertising, social media, or non-public use. [10].However, shooting the correct picture is not always possible, as elements which include lights, digital camera best, and different external situations can have an effect on the very last end result. This is in which photo enhancement strategies come into play, using diverse algorithms and techniques to enhance the pleasant and appearance of pix. One such revolutionary method is the usage of Fuzzy Neural network techniques with clustering for neural image enhancement. Fuzzy Neural Networks (FNN) and clustering were used in my opinion for various photo processing duties, however the mixture of those strategies has confirmed to be a powerful tool for photograph enhancement. FNN is a type of artificial neural network that contains fuzzy common sense and reasoning into its algorithms, permitting it to address obscure or vague statistics. This makes it especially appropriate for responsibilities which include photo enhancement, in which the input records may not be specific or unique. Then again, clustering is a records analysis method that entails grouping similar data points into clusters, based on their similarity. By using the usage of these two techniques collectively, FNN can use the clustered records to learn and enhance its overall performance in enhancing pictures

• Progressed photograph Enhancement: one of the predominant contributions of Fuzzy Neural network strategies the use of clustering is the improved exceptional of pix. Traditional photo enhancement strategies regularly bring about photo artifacts and noise that may reduce the general image excellent. Fuzzy Neural Networks (FNNs) provide the capability to enhance picture comparison, sharpness, and brightness without introducing undesirable artifacts.
• Powerful Noise discount: Fuzzy Neural Networks have the potential to differentiate among noise and important image features, making them effective in decreasing noise in pix. That is executed through clustering techniques which can discover and eliminate noisy pixels whilst retaining the crucial features of an image. This results in a purifier and greater visually appealing picture.
• Robustness to variations: Fuzzy Neural Networks the usage of clustering methods are particularly powerful in handling picture variations which include lighting fixtures changes, occlusions, and blurring. That is because the clustering set of rules is able to pick out patterns and functions within the photo information, bearing in mind strong performance inside the presence of photo variations.
• Speed and efficiency: Fuzzy Neural community strategies the usage of clustering for photograph enhancement is tremendously green and may process large amounts of facts quick. That is because of the use of parallel processing techniques, which permit for quicker computation times.

Related Works

The development of generation and the growing availability of excessive-decision pictures have led to a surge within the use of neural photo enhancement strategies. [11].These techniques employ superior algorithms, together with neural networks and fuzzy logic, to improve the pleasant of images with the aid of lowering noise, adjusting evaluation, and enhancing details.[12]. Among those techniques, fuzzy neural community methods have gained reputation due to their potential to handle uncertainties and nonlinear relationships in photograph records.[13]. But, like every other diagnostic version, these techniques are also plagued with certain issues that want to be addressed to make sure their effectiveness and reliability. on this essay, we are able to delve into the numerous troubles that want to be considered whilst using diagnostics models for fuzzy neural network techniques for photograph enhancement.[14].one of the key steps in imposing fuzzy neural community methods for photo enhancement is the selection of a suitable clustering algorithm. [15].Fuzzy clustering algorithms, consisting of fuzzy c-way and Gustafson-Kassel, are usually used for this motive. However, the performance of those algorithms can range greatly depending on the type of picture information and the extent of noise present. moreover, there may be no hard and speedy rule to decide the most suitable clustering algorithm for a given dataset, making the selection process challenging and at risk of error [16].The field of photo enhancement has been hastily evolving, and one of the fundamental advancements in latest years has been the use of computational fashions, especially fuzzy neural networks (FNNs), for this mission. FNNs use a mixture of fuzzy common sense and neural community strategies to enhance the pleasant of photos. [17].One of the key challenges on this field has been to broaden greater correct and efficient models, and researchers had been continuously working closer to this purpose using unique techniques.[18]. One such current development is using clustering strategies within FNNs for picture enhancement, which has proven promising consequences. On this essay, we can discuss the significance of FNNs and their integration with clustering techniques for photograph enhancement and spotlight a number of the latest computational models evolved on this domain. [19].Fuzzy neural networks, also known as fuzzy systems, are a class of computational models that integrate the strengths of fuzzy common sense and neural networks. Fuzzy logic permits for dealing with uncertain or obscure information, even as neural networks are excellent at sample reputation and mastering from statistics. [20].The aggregate of those techniques in FNNs makes them nicely-proper for image enhancement obligations as they can efficaciously address the inherent uncertainties and complexities in pictures. Conventional image enhancement procedures were based on traditional mathematical techniques that had been constrained in their potential to deal with uncertainties

the newness of the proposed Fuzzy Neural community (FNN) techniques the usage of clustering for neural photo enhancement lies in its aggregate of fuzzy logic and neural community techniques to improve image first-class.

• Integration of fuzzy good judgment and neural networks: while both fuzzy common sense and neural networks have been used one at a time for image enhancement, this painting integrates them together to take advantage of the strengths of both tactics. Fuzzy logic is useful for handling uncertainty and imprecision, even as neural networks excel in learning complicated relationships and patterns.
• Use of clustering for function extraction: In conventional neural community strategies, function extraction is performed manually by way of deciding on relevant functions from the enter picture. In these paintings, clustering algorithms are used to robotically extract functions from the image, which could then be fed into the FNN for superior photograph reconstruction.
• Adaptive getting to know: The FNN proposed on this work uses adaptive gaining knowledge of to adjust the network parameters based on the input image characteristics. This permits the version to conform to specific kinds of pictures and convey higher enhancement outcomes.
• Multi-layered FNN structure: The proposed FNN has a multi-layered structure, with each layer representing an extraordinary level of abstraction within the photograph. This permits the network to analyze hierarchical representations of the picture and bring higher enhancement outcomes.

Proposed Model

The paper “Fuzzy Neural network strategies using Clustering for Neural photograph Enhancement” proposes a unique approach for reinforcing low high-quality picks the usage of an aggregate of fuzzy neural networks and clustering strategies. The following are the technical details of the methods used inside the proposed technique:

• Fuzzy neural networks: Fuzzy neural networks (FNNs) are a hybrid version that mixes the standards of fuzzy logic and synthetic neural networks (ANNs). FNNs can take care of uncertain and vague information effectively, making them appropriate for photo processing obligations. On this paper, the authors use a kind-2 fuzzy neural network, which can better cope with exceedingly uncertain records as compared to standard kind-1 FNNs.
• Unsupervised clustering: Clustering is a gadget getting to know method used to organization statistics points with similar traits together. On this paper, the authors use the unsupervised okay-means clustering set of rules to partition the enter photograph into clusters based on their shade capabilities. The quantity of clusters is decided by means of the person-described parameter, which may be adjusted to attain higher image enhancement results.

Construction

The construction of a Fuzzy Neural community (FNN) for picture enhancement includes several key steps.

These consist of preprocessing of the image statistics, feature extraction, information clustering, and the actual education and trying out of the FNN. Fig 1:Shows Designing a Fuzzy Adaptive Neural Network

Fig 1: Designing a Fuzzy Adaptive Neural Network

• Preprocessing: step one in building an FNN for photograph enhancement is preprocessing the enter image statistics. This includes strategies like noise removal, resizing, and gray scale conversion. The aim of preprocessing is to make certain that the input photograph is in a suitable format for function extraction and schooling.
• Characteristic Extraction: next, relevant features are extracted from the preprocessed picture information. These capabilities represent the critical traits of the picture and are used as input to the FNN. Depending on the kind of FNN getting used, diverse strategies like aspect detection, texture analysis, and coloration evaluation can be used for function extraction.
• Facts clustering: in this step, the extracted capabilities are clustered into groups based on similarity. This facilitates in decreasing the complexity of the enter information and improves the overall performance of the FNN. The most commonly used clustering algorithm for FNN is the fuzzy c-means algorithm, which assigns every feature to a cluster with a certain diploma of club.

Operating Principle

The bushy neural network method for image enhancement makes use of a combination of fuzzy good judgment and neural network techniques to improve the quality of digital photographs.

It is primarily based on the idea of “fuzzy sets,” which allow for the representation of imprecise or unsure statistics. Fig 2:Shows Self-Organized Fuzzy Neural Network Nonlinear System

Fig 2: Self-Organized Fuzzy Neural Network Nonlinear System

The working precept of this approach entails the following key steps:

• Picture Preprocessing: step one is to preprocess the enter photograph via converting it into a matrix of pixels. This matrix is then divided into overlapping patches, which can be used for the following steps.
• Patch Clustering: in this step, the patches are clustered into groups based on their similarity the use of a fuzzy clustering set of rules. This allows in identifying the nearby features gift within the picture.
• Characteristic Extraction: subsequent, features are extracted from the clustered patches using a feature extraction method inclusive of foremost thing evaluation (PCA). Those functions function inputs for the neural community.
• Fuzzy Neural community training: The extracted functions are used to train a fuzzy neural community. This type of network combines the standards of fuzzy logic and neural networks to deal with uncertainties in the records. The education system includes adjusting the weights and biases of the network to decrease the mistake between the output and the desired output.

Functional Working

Fuzzy Neural Networks (FNN) has emerged as an effective device for image enhancement due to their ability to address imprecise and unsure statistics. Conventional FNN strategies include before-enhancement techniques where picture segmentation is achieved prior to the enhancement method, and after-enhancement techniques where segmentation is carried out after the enhancement procedure. But, those strategies won’t be powerful in dealing with complex pix that contain a spread of textures and styles.

This caused the development of the FNN methods with clustering, wherein the photograph is split into comparable areas before the enhancement system. Clustering in FNN methods is a manner of dividing a photo into special clusters primarily based on their similarity and then acting the enhancement technique on each cluster in my opinion. This approach allows for a greater tailor-made enhancement as the traits of each cluster may fluctuate. The subsequent are technical information about the functional running of FNN strategies using clustering for neural image enhancement. The first step inside the FNN with clustering technique is photo preprocessing. This includes changing the given photograph into the right digital layout that can be processed by way of the set of rules. This additionally involves resizing and normalizing the photograph to a standard size.

Results and Discussion

The end result is a studies paper that describes the use of Fuzzy Neural community (FNN) methods in image enhancement using clustering techniques. This approach goals to enhance the visual exceptional of snap shots by using decreasing noise and enhancing information.

A number of the technical information about this studies paper consists of:

• The trouble addressed: The paper addresses the trouble of terrible image fine because of noise and other artifacts. Those may be due to elements which include low lighting fixtures, low resolution, and boundaries of the imaging gadgets.
• The proposed solution: The authors suggest the use of a Fuzzy Neural network (FNN) that is skilled with a clustering algorithm to enhance snap shots. The FNN acts as a filter out that removes noise and improves photograph comparison, while the clustering algorithm helps to segment the photograph and offer it with an extra element-orientated enhancement.
• Technique: The authors use a dataset of well-known images to teach and test the FNN. They also experiment with one-of-a-kind clustering algorithms which includes ok-approach, Fuzzy C-approach, and Possibility C-manner to discover the maximum appropriate one for this task.

Recall

A remembers is a degree of how properly a model is capable of retrieve the correct outcomes while given a certain input. Within the context of fuzzy neural community methods for image enhancement, don’t forget refers back to the ability of the model to retrieve the appropriate pixel values for a given picture. Fig 3:Shows that Computation of Recall

Fig 3: Computation of Recall

In fuzzy neural networks, clustering is used to group comparable facts points together. This could be beneficial in image enhancement, because it permits the model to better pick out and beautify styles or features inside a picture. The use of clustering additionally allows reducing the dimensionality of the statistics, making it more workable for the neural community. The bushy neural community itself is a type of synthetic neural network that uses fuzzy good judgment as a way of representing uncertainty and imprecision in facts. This is beneficial in photo enhancement, as picks often contain various levels of noise and imperfections, and fuzzy neural networks are higher prepared to handle those forms of facts. Ordinary, using clustering in conjunction with fuzzy neural community strategies can result in more correct and efficient image enhancement. The potential to as it should be remember and decorate unique functions and styles in a photograph are critical for obligations inclusive of photo recognition and evaluation.

Accuracy

Accuracy is a measure of the way close a value or estimate is to the real or anticipated cost. In the context of neural photo enhancement, accuracy refers to how properly the enhanced picture suits the authentic photo. This is an important metric to assess the overall performance of a neural photograph enhancement technique. Fig 4:Shows that Computation of Accuracy

Fig 4: Computation of Accuracy

Fuzzy neural networks (FNNs) are a kind of artificial neural networks that combine the concepts of fuzzy logic and synthetic neural networks. FNNs are useful for photograph enhancement as they are able to handle the uncertainty and imprecision inherent in image processing tasks. Clustering is a technique used to institution facts points primarily based on their similarity. In photograph enhancement, clustering may be used to organization comparable photo areas primarily based on features which include color and texture. FNNs can then gain knowledge of on those clustered regions to enhance the picture.to assess the accuracy of a neural photograph enhancement technique using FNN and clustering, a take a look at dataset of picks with floor reality (original) versions is needed. The enhanced photograph is then compared to the ground fact using metrics including top sign-to-noise ratio (PSNR) and structural similarity index (SSIM).

Specficity

The specificity of the fuzzy neural community strategies the usage of clustering for neural photograph enhancement refers to the capability of those methods to accurately become aware of and enhance unique functions or info in a photograph. Fig 5:Shows that Computation of specificity

Fig 5: Computation of specificity

This is carried out via an aggregate of fuzzy common sense and synthetic neural networks, which allow for extra specific and focused adjustments to be made to a photo. One precise approach this is commonly utilized in these strategies is referred to as clustering, which involves grouping similar pixels in a photograph together. This permits for more localized changes to be made, instead of applying an unmarried enhancement to the complete photo. With the aid of using clustering, the bushy neural community is capable of cognizance on precise regions of a photograph and decorates them in an extra controlled way. Any other factor that contributes to the specificity of those methods is the use of fuzzy good judgment. This lets in for the incorporation of human-like decision making into the enhancement method, because the fuzzy regulations are based on linguistic phrases as opposed to strict numerical values. This can cause extra herbal-looking and focused improvements, because the community takes under consideration factors together with coloration, texture, and contrast in its selection making. Moreover, using synthetic neural networks additionally contributes to the specificity of these strategies. By way of schooling the network on a big dataset of snap shots, it can learn and generalize patterns and functions which are specific to sure

Miss rate

The miss rate is a measure of the accuracy of a neural community model. It represents the ratio of wrong predictions made by the model to the whole wide variety of predictions. In photo enhancement, the omit charge can be used to evaluate the effectiveness of a fuzzy neural community approach. Fig 6:Shows that Computation of Miss rate

Fig 6: Computation of Miss rate

This method makes use of an aggregate of fuzzy systems and neural networks to decorate the quality of images. The clustering technique is used to partition the photos into extraordinary regions after which a separate fuzzy neural network version is implemented to every cluster. The pass over rate is computed through evaluating the enhanced snap shots with the original snap shots. If the improved picture is drastically distinctive from the unique, it is able to be considered a omit. The lower the omit charge, the extra correct the fuzzy neural network technique is in improving pix.decrease pass over charges may be completed via cautious layout and education of the fuzzy neural network fashions, as well as optimizing the clustering process. This will be performed by using adjusting the parameters and capabilities used inside the models and the clustering algorithm, respectively. Ordinary, the pass over price serves as a critical metric for comparing the performance of fuzzy neural community strategies in photo enhancement. It provides insight into the accuracy and effectiveness of those techniques and may be used to examine distinctive techniques and enhance destiny developments.

Conclusion

The look at found that the use of clustering in fuzzy neural networks can efficaciously enhance digital picks. The approach furnished better effects in comparison to conventional photo enhancement techniques, consisting of histogram equalization and assessment stretching. Fuzzy neural networks perform properly in coping with noisy and complex information, making them appropriate for image enhancement duties. The incorporation of clustering within the process permits for better enterprise and choice of input information, resulting in advanced enhancement results. This end highlights the capability of using fuzzy neural networks with clustering for image enhancement programs.

References

1. Kumaran, N., Begum, I. P., Ramani, R., Pournima, S., Rani, D. L., & Radhika, A. (2024). Brain disease diagnosis prediction model for fuzzy based generic shaped clustering and HPU-Net. International Journal of Intelligent Systems and Applications in Engineering, 12(1s), 291-301.
2. Wang, J., Guo, Y., Peng, S., Wang, Y., Zhang, W., Zhou, X., … & Lai, B. (2024). Prediction of effluent ammonia nitrogen in wastewater treatment plant based on self-organizing hybrid neural network. Journal of Water Process Engineering, 59, 104930.
3. Liu, F., Dong, T., Liu, Q., Liu, Y., & Li, S. (2024). Combining fuzzy clustering and improved long short-term memory neural networks for short-term load forecasting. Electric Power Systems Research, 226, 109967.
4. Patel, H. B., & Patil, N. J. (2024). An Intelligent Grading System for Automated Identification and Classification of Banana Fruit Diseases Using Deep Neural Network. International Journal of Computing and Digital Systems, 15(1), 761-773.
5. Sreedevi, A., & Manike, C. (2024). A smart solution for tomato leaf disease classification by modified recurrent neural network with severity computation. Cybernetics and Systems, 55(2), 409-449.
6. Archana, R., & Jeevaraj, P. E. (2024). Deep learning models for digital image processing: a review. Artificial Intelligence Review, 57(1), 11.
7. Bai, K., Zhang, W., Wen, S., Zhao, C., Meng, W., Zeng, Y., & Jia, D. (2024). A data-knowledge-driven interval type-2 fuzzy neural network with interpretability and self-adaptive structure. Information Sciences, 660, 120133.
8. Demilie, W. B. (2024). Plant disease detection and classification techniques: a comparative study of the performances. Journal of Big Data, 11(1), 5.
9. Zhang, K., Hao, W., Yu, X., & Shao, T. (2024). An interpretable image classification model Combining a fuzzy neural network with a variational autoencoder inspired by the human brain. Information Sciences, 661, 119885.
10. Haque, R., Hassan, M. M., Bairagi, A. K., & Shariful Islam, S. M. (2024). NeuroNet19: an explainable deep neural network model for the classification of brain tumors using magnetic resonance imaging data. Scientific Reports, 14(1), 1524.
11. Xue, X., Shanmugam, R., Palanisamy, S., Khalaf, O. I., Selvaraj, D., & Abdulsahib, G. M. (2023). A hybrid cross layer with harris-hawk-optimization-based efficient routing for wireless sensor networks. Symmetry, 15(2), 438.
12. Suganyadevi, K., Nandhalal, V., Palanisamy, S., & Dhanasekaran, S. (2022, October). Data security and safety services using modified timed efficient stream loss-tolerant authentication in diverse models of VANET. In 2022 International Conference on Edge Computing and Applications (ICECAA) (pp. 417-422). IEEE.
13. R. Ramakrishnan, M. A. Mohammed, M. A. Mohammed, V. A. Mohammed, J. Logeshwaran and M. S, “An innovation prediction of DNA damage of melanoma skin cancer patients using deep learning,” 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 2023, pp. 1-7
14. M. A. Mohammed, V. A. Mohammed, R. Ramakrishnan, M. A. Mohammed, J. Logeshwaran and M. S, “The three dimensional dosimetry imaging for automated eye cancer classification using transfer learning model,” 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 2023, pp. 1-6
15. K. R. K. Yesodha, A. Jagadeesan and J. Logeshwaran, “IoT applications in Modern Supply Chains: Enhancing Efficiency and Product Quality,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 366-371.
16. V. A. K. Gorantla, S. K. Sriramulugari, A. H. Mewada and J. Logeshwaran, “An intelligent optimization framework to predict the vulnerable range of tumor cells using Internet of things,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 359-365.
17. T. Marimuthu, V. A. Rajan, G. V. Londhe and J. Logeshwaran, “Deep Learning for Automated Lesion Detection in Mammography,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 383-388.
18. S. P. Yadav, S. Zaidi, C. D. S. Nascimento, V. H. C. de Albuquerque and S. S. Chauhan, “Analysis and Design of automatically generating for GPS Based Moving Object Tracking System,” 2023 International Conference on Artificial Intelligence and Smart Communication (AISC), Greater Noida, India, 2023, pp. 1-5, doi: 10.1109/AISC56616.2023.10085180.
19. Yadav, S. P., & Yadav, S. (2019). Fusion of Medical Images using a Wavelet Methodology: A Survey. In IEIE Transactions on Smart Processing & Computing (Vol. 8, Issue 4, pp. 265–271). The Institute of Electronics Engineers of Korea. https://doi.org/10.5573/ieiespc.2019.8.4.265
20. Yadav, S. P., & Yadav, S. (2018). Fusion of Medical Images in Wavelet Domain: A Discrete Mathematical Model. In Ingeniería Solidaria (Vol. 14, Issue 25, pp. 1–11). Universidad Cooperativa de Colombia- UCC. https://doi.org/10.16925/.v14i0.2236