Dr Sepehr Ghaffari

One of the major benefits of the pre-stressed concrete pavements is the omission of tension in concrete that results in a reduction of cracks in the concrete slabs. Therefore, the life of the pavement is increased as the thickness of the slabs is reduced. One of the most important issues in dealing with the prestressed concrete pavement is determination of the magnitude of the pre-stress. Three dimensional finite element analyses are conducted in this research to study the pre-stress under various load (Boeing 777) and thermal gradient combinations. The model was also analyzed under temperature gradients without the presence of traffic loading and the induced stresses were compared with those from theoretical relationships. It was seen that the theoretical relationships result in conservative values for the stress.

Fracture properties and crack propagation characteristics of asphalt concrete mixtures were studied by obtaining fracture resistance curves using three point single edge SE(B) notched beam specimens. Elastic-plastic approach is used in the calculation of the J-integral since the fracture process zone size is large enough to not use a linear elastic approach. Crack length measurements were obtained directly from high resolution images taken during the tests. A rising R-curve was observed in all the specimens which indicates ductility and a toughening mechanism in the ductile to quasi-brittle fracture of the mixture. Mixtures developed by limestone and siliceous aggregates with 4%, 4.5% and 5% binder contents were tested at temperatures ranging from +5ºC to –20ºC. Mixtures with 5% binder content showed greater crack resistant behavior at each temperature. Crack lengths at which crack propagation instability occurred were decreased by the reduction of temperature. A significant drop of this critical crack length is observed in temperatures below –15ºC. As well, the elastic-plastic fracture toughness is increased by the reduction of temperature up to –15ºC and starts to diminish thereafter.

A single specimen test using the three point single edge notched beam configuration at low temperatures for obtaining hot mix asphalt (HMA) resistance curves is developed. Resistance curves are obtained for mixtures at six temperature levels of +5, 0, −5, −10, −15, and −20 °C and three binder contents of 4%, 4.5%, and 5%. Crack extension increments during the test are measured by means of an image processing technique using Radon transform and feature extraction. All the specimens exhibit a rising R-curve, indicating ductility and toughening mechanisms in the ductile-quasi brittle fracture of the mixture. It is observed that the reduction of temperature results in a further tendency of the mixture for unstable crack growth and less subcritical crack length. It is also shown that using the binarization process, an automatic index can be developed that can represent the extent of brittleness and extent of the low temperature in which the cracking has occurred.

In this research, a sustainable approach is followed to develop efficient mixtures incorporating recycled fine aggregate (RFA) remained from structure demolition as well as limestone filler (LF) from production of hot mix asphalt (HMA). The LF is a byproduct of the drying process in HMA production plant which is not entirely consumed in the production of the HMA and must be hauled and disposed in landfills. The maximum particle size of the LF is approximately 40 µm. Self-Compacting Concrete (SCC) mixtures were designed replacing 5% and 10% of the cement with LF. Incorporation of 50%, and 100% RFA with the fines in the mixtures were considered with and without addition of the LF. Due to the formwork and prefabrication restrictions, the paste volume and the high range water reducer content were tuned in such a way that the slump flow of the mixtures remained between 660 mm to 700 mm without segregation. Durability and mechanical performance of the mixtures were evaluated by resistance against freeze-thaw scaling exposed to deicing agents and compressive strength. It was observed that the SCC mixtures containing 10% LF outperformed those without the use of LF while 5% SCC mixtures did not exhibit tangible superiority. Incorporation of RFA as the fine fraction degraded the durability of all the mixtures. While replacing all the fine fraction with RFA significantly impaired durability and compressive strength, 50% RF mixtures could be designed containing 10% LF that remained in the allowable limits.

In a previous research by authors, a methodology was developed to derive J-R curves for Hot Mix Asphalt (HMA) mixtures using an elastic-plastic approach where a comprehensive understanding of crack propagation regime could be achieved. In this research, the effect of crumb rubber modification of HMA binder is studied in terms of R-curves and crack propagation at low temperatures. Mode I Single edge notched beam (SE(B)) fracture tests were conducted in temperature levels of 0 °C, -10 °C, and -20 °C. PG58-22 and PG64-22 binders were used in the fabrication of HMA samples. Modified specimens consist of 20% crumb rubber along with the incorporation of 3% warm mix admixture. Crack growth resistance curves were obtained in SE(B) tests by means of image processing and recording of the progressive crack length. Elastic-plastic J-R curves revealed that crumb rubber modified mixtures exhibit higher crack growth resistance for each bitumen performance grade. As well, increased ductility and cohesive energy can be observed according to the R-curves as the mixtures are modified by crumb rubber.

R-curves provide the means for in-depth analysis of crack propagation in engineering materials. However, due to experimental costs and difficulties, the application of this approach has been very limited in hot mix asphalt (HMA) investigations to date. In the present research, single edge notched beam (SE(B)) tests are conducted on asphalt concrete (AC) mixtures in pure mode I and mixed mode (I/II) states at low temperatures. Mode mixity is induced in the test setup by fabricating an asymmetric notch with varying relative offset values of 0.3, 0.47, and 0.67 from the centerline of the beam where the loading is situated. Mode I and mixed mode R-curves were constructed by plotting cumulative fracture energy versus crack extensions obtained from digital images during each test. Based on the R-curves, the mixed mode (I/II) crack propagation regime of the HMA with regards to temperature variations (+5 °C, 0 °C, −5 °C, −10 °C, −15 °C, and −20 °C in this research), binder contents (4%, 4.5%, and 5%), and the nominal maximum aggregate size (NMAS) of 19 mm and 25 mm were determined and studied. A considerable stable crack growth zone was observed in the mixtures as the contribution of shearing mode is increased and the mode I contribution is decreased. The stable crack growth region was proceeded by unstable crack propagation in all the cases and the instability outweighed in lower temperatures, particularly in −15 °C and −20 °C. Nevertheless, the mixtures exhibited greater resistance to post-peak crack propagation in higher mode mixites which could be inferred from the progressively rising R-curve in the instability phase. It was also observed that the NMAS value of 25 mm could increase the cohesive energy and the blunting work required to initiate macrocracking while amplifying unstable crack growth at −20 °C and −15 °C in comparison with NMAS 19 mm.

J-Resistance curves serve as a robust tool in characterizing the entire crack propagation trend of materials. Nevertheless, thus far, investigations on J-R curves of hot mix asphalt (HMA) have been very limited leading to scarce knowledge of post-peak fracture of the mixtures. In this research, mode I single-edge notched beam (SE(B)) fracture testing of unmodified and crumb rubber (CR) modified asphalt concrete mixtures was conducted with continuous monitoring of crack extension during the tests at 0 °C, −10 °C, and −20 °C. Modified HMA beams were fabricated incorporating 10% and 20% CR with 3% warm-mix additive. The CMOD-based elastic–plastic J-integral was computed incrementally during the tests and the J-R curve of each mixture could be determined. The results indicated that reducing the temperature increases the crack blunting energy from 0 °C to −10 °C while reducing the extents of the transition zone within blunting to unstable crack propagation of the unmodified mixtures. The unmodified mixtures undergo an unstable crack propagation as early as the macro crack initiates in −20 °C. At the same time, the incorporation of CR, could increase the blunting fraction of the R-curves at each temperature. Of vital importance, it could be detected that a 20% crumb rubber modification resulted in a more significant stable crack growth zone in the mixtures while developing a progressively rising R-curve in the post-peak region than the flat state in the unmodified mixtures at severely low temperatures.

Fracture resistance curves (R-curves) provide a robust tool for a comprehensive insight into the crack propagation regime in engineering materials. In this paper, an extensive research program is conducted to determine R-curves for hot mix asphalt (HMA) mixtures with varying properties. The experimental results are then used to develop R-curve prediction models following a machine learning approach. Three-point single-edge notched beam (SE(B)) experiments were conducted on HMA mixtures incorporating 0%, 5%, 10%, 15%, and 20% crumb rubber at low temperatures. The temperature ranged from + 5 °C to −20 °C while limestone and siliceous aggregate with two gradations were used in developing mixtures with two base bitumen having performance grades of PG58-22 and PG64-22. It was observed that as the temperature is declined to −20 °C, the stable crack growth region is significantly diminished in the R-curves, and the mixtures undergo a brittle fracture with abrupt failure of the specimen. A temperature of −15 °C could be determined where the transition from quasi-brittle to brittle fracture occurs. Mixtures fabricated incorporating 20% crumb rubber exhibited a progressively rising R-curve at the lowest test temperature (−20 °C) even in the unstable crack propagation phase, which is a desirable material characteristic. Two prediction models were developed for R-curves. Artificial neural networks (ANN) were used in the first model resulting in an R-square value of 0.965. Due to the black-box nature of the ANN, the multi-gene genetic programming approach was also applied in the prediction of the R-curves to derive a mathematical equation between the input data and the outputs. The R-square equaled 0.870 in this method. R-curves could successfully be predicted by both methods considering the negligible to fair errors.

Determining mixed mode fracture parameters asphalt concrete mixtures remains an engineering challenge due to non-homogeneity and inelasticity of the material. In this research, a study was conducted to determine the low-temperature R-curves of unmodified and crumb rubber modified Hot Mix Asphalt (HMA) under mode I and mixed-mode (I/II) loading conditions. Single edge notched beam (SE(B)) testing was employed to collect data, and three key fracture parameters—cohesive energy, energy rate, and fracture energy—were extracted to represent different stages of fracture and crack propagation. Within the scope of this study, it was observed that for the AC 85/100 paving grade bitumen, a temperature of − 20 °C serves as a critical temperature, shifting fracture from quasi-brittle to brittle. At this temperature, the stable crack growth region in the R-curves significantly shrinks, causing abrupt specimen failure. The incorporation of 20% crumb rubber demonstrated favorable material characteristics, with a progressively rising R-curve even during the unsfi crack propagation phase. The central goal of this research is to establish prediction models for the mixed-mode (I/II) crack propagation parameters Gb, Gf, and Gi. The features selected for modeling are Gb0, Gf0, and Gi0 (mode I), percentage of crumb rubber, type of aggregate, binder content, nominal maximum aggregate size, temperature, and normalized offset ratio. Two dataset configurations were used: dataset 1 contains all entries, while dataset 2 excludes Gb0, Gf0, and Gi0 (mode I). Five machine learning techniques, Regression, Multi-Gene Genetic Programming (MGGP), Support Vector Regression (SVR), Random Forest, and Artificial Neural Networks were employed to predict three key fracture parameters. Although slightly less accurate than SVR and Random Forest, MGGP offers the key advantage of yielding explicit mathematical expressions for crack propagation prediction. The R2 index for the MGGP model in Dataset 1 was 0.93 for Gb, 0.94 for Gf, and 0.92 for Gi. For dataset 2, the indices were 0.89, 0.93, and 0.88, respectively.

Traditional pavement repair techniques are time-consuming, labour-intensive, prone to errors, and expose manpower to high-risk road traffic conditions. This paper proposes a data-driven solution for planning and automating the repair process for road potholes using a fleet of unmanned ground vehicles (UGVs). The project encompasses data mining, developing software tailored for fleet management, and enhanced fault tolerance. Additionally, it incorporates the integration of digital twins for advanced simulation purposes. The methodologies involve cross-industry standard processes for data mining (CRISP-DM) and preparation combined with rapid application development (RAD). To optimise repair schedules, the system takes parameters like fleet size, payload capacity, and material requirements based on pothole dimensions. This data-driven project concludes from simulations that a neighbourhood can be patched about 40 % faster and optimised to achieve a 12.5 % reduction in robot inter-travel time using three UGVs per defined residential area of 100,000 m2 instead of two UGVs in the fleet.

In this study, asphalt concrete (AC) mixtures were modified with polymerized sulfur, using PG58-22 bitumen, and crushed siliceous aggregate. Modifications involved replacing the base binder with 20%, 30%, and 50% polymerized sulfur, compared to a control mix with no replacement. The mixtures were subjected to Single Edge Notched-Beam (SE(B)) fracture tests under mixed mode (I/II) conditions with notch offset value of 48 mm, with temperatures ranging from 0 °C to -20 °C. These tests, focusing on the mixtures' response to mixed mode loading, provided load-displacement curves, enabling the determination of fracture energy. Results indicated an increase in fracture energy for 20% and 30% sulfur-modified mixtures. However, a trend towards increased embrittlement was also observed, as fractures occurred at lower displacements. Significantly, higher sulfur content correlated with similar or decreased mixed-mode (I/II) fracture energy, suggesting an improved resistance to low-temperature cracking for lower replacement percentages.

In recent years, there has been a growing emphasis on incorporating recycled materials into the production of eco-friendly construction materials. Additionally, the substantial disposal of end-of-life tires poses significant environmental challenges. Using waste tire constituents in the geopolymer materials as binders a feasible and sustainable alternative to conventional concrete with Portland Cement, effectively resolving environmental problems and promoting the advancement of eco-friendly building material production. This research focuses on producing eco-friendly alternatives to Portland Cement-based products. Moreover, it is aimed to explore the utilization of all components of the end-of-life tires (crumb rubber, textile, and steel fibers) in the mixture and replace Portland Cement with a geopolymer matrix in mortar mixtures. To achieve this, recycled waste tire materials were added to the geopolymer mixtures based on the design of experiments using a central composite design approach. An RSM-based assessment was applied to determine the effect of incorporating recycled materials on the durability and mechanical properties of the geopolymer mixtures. Moreover, the environmental impacts of mixtures were evaluated based on global warming potential and gross energy requirements. The analysis of experimental results revealed that incorporating recycled materials not only substantially mitigated the environmental impact but also improved mixture properties, i.e. reduced water absorption, enhanced resistance to salt scaling, and increased impact resistance. However, there was a slight decrease in compressive strength. Considering all aspects, the final optimized mixture saved 1172.77 MJ/ton energy and reduced 52.33 kg CO2eq/ton per 1 m3 compared to the mixture without recycled materials.

Mixed mode (I/II) loading conditions occur frequently in the asphalt layers of pavements. Therefore, a low-temperature fracture analysis based on mixed mode loading turns out to be of utmost importance. In this research, asphalt concrete (AC) mixtures were prepared using two aggregate gradations and PG58-22 bitumen. AC beams were produced by the mixtures and notch offset values of 48 mm, 75.2 mm, and 107.2 mm were fabricated in the beams in order to be tested in a modified single-edge notched beam (SE(B)) setup. The tests were carried out at two temperature levels of-5 °C and-15 °C. Using the modified SE(B) setup and capturing and processing digital imaged from the growing crack during the tests, fracture resistance curves (R-curves) in mixed mode (I/II) conditions could be constructed for each mixture. The results revealed that increasing the mode mixity and impairing the tensile mechanism in the fracture of asphalt beams could significantly contribute to higher fracture resistance of the mixtures. Mixtures with the highest mode mixity exhibited greater crack tip blunting energy by up to 25%. Similarly, energy dissipation in the unstable crack propagation zone is also increased being a desirable characteristic in post-peak performance of the mixtures.

Fracture regime of asphalt concrete is of utmost importance in the sustainable design of optimized mixtures against low-temperature cracking. The energy dissipated in blunting the crack tip, stable crack growth, as well as post-peak resistance of the mixtures to the propagating crack comprise the overall fracture performance of the mixtures. In this research, employing the fracture resistance curve (R-curve) concept, three energy parameters: cohesive energy, fracture energy, and energy rate, are extracted to quantify the crack propagation regime of the mixtures incorporating ground recycled waste tire. A temperature range of 0 °C to −20 °C with 0 %, 10 %, and 20 % crumb rubber contents (CRC) were considered and the experiments were carried out in Single Edge notched Beam (SE(B)) fracture testing protocol. An environmental index comprising CO2 and CH4 emissions, as well as the energy consumption, was developed and the production cost of AC samples were also determined. A design of experiments based on the Central Composite Design (CCD) is applied. The Response Surface Methodology (RSM) is used to develop the best model between the fracture response, environmental factor, cost, and the input mixture properties. Multi-objective optimization scenarios were assessed by the RSM and a 4.74 % binder content with 19.5 % crumb rubber incorporation was resulted as the optimal mix design for maximum fracture response and minimized cost and environmental concerns in low temperatures. The environmental effects and costs were by average reduced by 15 % and 1 %, respectively and fracture responses have been improved compared to the reference optimum mix design (CRC = 0).

This study focuses on enhancing the predictive accuracy and generalization of artificial neural networks (ANN) for the compressive strength of self-compacting concrete (SCC), utilizing datasets that mainly feature properties of conventional concrete. A dataset comprising 1100 sets of data was blended with an originally developed dataset obtained from compressive strength testing of SCC mixtures. SCC mixtures were produced using 3.19 to 6.12 lit/m3 of high range water reducer to attain a target slump flow range of 550 mm to 700 mm. Compressive strength testing was carried out for all the mixtures at ages of 7, 14, 28, and 90 days. Eight input variables were selected for modeling, including cement, water, fine and coarse aggregate, superplasticizer, fly-ash content, and age. In the first stage, a simple ANN model, serving as a baseline for performance comparison was developed. This model yielded an R-squared value of 0.84, demonstrating a modest capability in predicting compressive strength of the SCC mixtures. Next, a deep ANN architecture was developed in the next stage with additional hidden layers and neurons which provided a marginal improvement in predictive accuracy increasing the R-squared from 0.84 to 0.87. Similar to the simple ANN, the deep ANN failed to predict the actual strength gain trend of the mixtures by age. Finally, implementing a systematic hyperparameter optimization across the deep model, a notable increase in the R-squared value, elevating it to 0.92, was achieved and the strength gain trend could be seamlessly predicted.

This study develops and characterizes a patented eco-friendly engineered cementitious composite (ECC) that incorporates waste glass powder (WGP) and silica fume (SF) as supplementary cementitious materials (SCMs) and recycled glass aggregate (WGA) as an alternative aggregate. Four stages of experimental design produced 14 concrete mixtures tested at 7, 28, 56, 90, and 120 days. Fresh and hardened properties were evaluated, and the optimal mixture, S8-1, A, achieved the requirements of strength class C60/75 and workability with slump class/ consistency class S3. Microstructural analyses using X-ray diffraction and optical microscopy confirmed the formation of secondary hydration products, particularly C-S-H and A-S-H, which contributed to matrix densification and improved performance. To complement the experimental program, an artificial neural network (ANN) was developed to predict compressive strength based on mixture proportions and curing age. Each strength measurement was treated as an independent data point, resulting in 70 samples for model training and testing. A shallow feedforward ANN with three hidden layers was implemented, trained using the Adam optimizer and validated with 10-fold cross-validation. The model achieved high predictive accuracy with R² of about 0.968, mean absolute error of 1.94 MPa, and root mean square error of 2.52 MPa. The results confirm that recycled WGP and SF can be effectively incorporated into ECC while ANN modeling provides a reliable tool for predicting compressive strength and supporting sustainable concrete mix design.

A modular C program to simulate the stress response of viscoelastic materials under arbitrary 1-D strain histories, using the generalized Maxwell model (Prony series). Users: Researchers and engineers in materials/structural mechanics University instructors for demonstrations/assignments Students learning viscoelasticity Outcome: Accurate and efficient solver with CSV I/O Extensible modular codebase Analysis tools for creep, relaxation, and dynamic mechanical analysis (DMA)

A modular implementation of a feedforward neural network (multi-layer perceptron) in C for classifying 28×28 grayscale images of handwritten digits (0-9).

A comprehensive C-based sensor data logging system designed for embedded systems, IoT applications, industrial monitoring, and civil engineering instrumentation.

Open-source C code written to calculate the elastic-plastic J-integral using its contour integral definition.

The global food industry prioritizes the quality and safety of fish and seafood products due to their perishable nature and the increasing consumer demand for nutritious, high-quality protein sources. Traditional quality assessment methods—such as sensory evaluation, chemical analysis, physical testing, and microbiological testing—form the foundation of current practices but face notable limitations, including subjectivity, destructiveness, and labor-intensive procedures that delay results. Motivated by the need for faster, more reliable, and nondestructive quality control systems, this article investigates how emerging technologies can address these limitations. Specifically, it aims to answer the following review questions: 1) what are the limitations of traditional fish quality assessment methods? 2) how can hyperspectral imaging (HSI) and computer vision improve the accuracy and efficiency of quality assessment? and 3) what roles do machine learning (ML) and deep learning (DL) techniques play in enhancing these technologies? This article explores the integration of HSI and computer vision as cutting-edge, noninvasive technologies enabling real-time, comprehensive evaluation of key fish quality attributes such as freshness, safety, nutritional content, and species verification. The fusion of HSI and computer vision with advanced learning algorithms enhances precision in quality control, reduces food waste, and supports compliance with modern standards. Finally, this article underscores the need for continued research to drive sustainable innovation and strengthen consumer confidence.

In this research, asphalt concrete (AC) mixtures modified by polymerized Sulfur were prepared. PG58-22 bitumen was used as the base binder for the mixtures along with crushed siliceous aggregate. The base binder was replaced by 20%, 30%, and 50% ratios with polymerized Sulfur in the modified mixtures while the reference mix was fabricated with 0% binder replacement. Single edge notched-beam fracture tests (SE(B)) were carried out in a temperature range of 0 °C to -20 °C on the AC beam specimens. Load-displacement curves were obtained from the experiments and the fracture energy of the mixtures could be determined. It was revealed that modifying the mixtures with polymerized Sulfur could improve the load bearing of the beam specimens as higher peak load values were recorded at fracture. However, fracture failure of the AC beams occurred at lower values of displacement addressing further embrittlement of the mixtures due to replacement of the base binder. Higher contents of polymerized Sulfur in the mixtures resulted in higher magnitudes of fracture energy as a general trend in this research addressing an improved resistance to low-temperature cracking.

Fracture resistance curves (R-curves) have served as a robust tool in characterizing the entire fracture process of engineering materials. However, obtaining such curves for asphalt concrete (AC) mixtures is cumbersome due to the non-linear inelastic behavior of the mixtures. In this research, a singlespecimen technique is developed based on the unloading compliance method which is used for metals. AC mixtures with limestone aggregate and PG58-22 binder were prepared. Beam specimens were fabricated and single-edge notched beam (SE(B)) fracture testing was conducted at low temperatures. A loading-partial unloading regime was used in the experiments and crack growth increments were captured by digital images throughout the tests. Using a multi-variable regression analysis, modified compliance equations were obtained for AC and R-curves of the mixtures could be constructed. It was revealed that the R-curve developed by ASTM E1820 compliance method could potentially overestimate the resistance of the mixtures against low-temperature fracture. The constructed R-curve exhibits a lower semi-vertical region addressing lower resistance of the mixture in the crack blunting phase. Also, the post-peak phase of the fracture shows a significantly lower slope in the constructed R-curve which denotes lower resistance of the mixture against unstable crack propagation.

As materials undergo large-scale yielding or exhibit large sizes of fracture process zone in the crack tip region, multi-parameter fracture concepts should be employed to describe the complex crack-tip stress-strain fields. Fracture resistance curves (R-curves) are an established tool in characterizing the entire fracture process of such materials. However, for complex materials such as bituminous mixtures, the development of these curves is subject to experimental and computational intricacies. In this research, a framework is developed to automate the construction of R-curves for normal and rubberized asphalt concrete (AC) mixtures. AC mixtures are produced using PG58–22 and PG58–28 binders. Limestone and siliceous aggregates are used, and three binder contents are considered for the mixtures. Single-edge notched beam (SE(B)) fracture testing is conducted on AC beams with two different notch patterns. A convolutional neural network (CNN) model is developed and trained over 1260 test images with varying temperatures, notch geometries, and setups. The CNN model was used to detect the growing crack on the beam surface and each crack-detected image was sent to an image processing framework to measure the crack length. Crack extension increments were calculated and synchronized with test time and magnitude of load, load-line displacement, and cumulative fracture energy, and the R-curve could be constructed. A training accuracy of 0.91 was obtained for the model and a loss of below 0.10 as a result of a hyperparameter tuning indicating reliable classifications by the CNN architecture. The R-curves showed desirable agreement for control mixtures at temperatures of 0 °C and −15 °C. As the mixtures are rubberized, the R-curves showed favorable agreement in the crack blunting phase, transition zone, as well as the unstable propagation phase at −20 °C. Cohesive energy magnitudes were compared for the two methods with a Pearson coefficient of 0.81 while fracture rate and fracture energy magnitudes showed favorably close magnitudes with coefficients of 0.89 and 0.98 respectively.

Source Camera Identification (SCI) is essential in digital image forensics, enabling reliable attribution of images to their originating devices for legal, investigative, and security applications. Yet, existing SCI methods often struggle under diverse imaging conditions due to scene-dependent noise and texture interference. This thesis advances SCI through four key contributions. First, a systematic evaluation of forty-two wavelets using the VISION dataset (34,427 images, 35 camera models, 11 brands) identified cdf9/7 as the most effective for Sensor Pattern Noise (SPN) extraction, followed by sym2 and coif1. Second, an Improved Camera Source Identification using Wavelet Noise Residuals and Texture Filtering (ICSI-WNRTF) method was developed to suppress high-texture regions, achieving 99% accuracy for model identification and 98% for device attribution. Third, a Curvelet-Based Camera Source Identification Leveraging Image Smooth Regions (CBCSI-SR) framework was introduced. By exploiting multi-scale directional features and isolating smooth regions, it achieved 99.6% model-level and 98.9% device-level accuracy while reducing false decisions. Finally, a Deep Learning-Based Texture Exclusion for Source Camera Identification (DLTESCI) approach combined texture suppression with a fine-tuned ResNet50, reaching 99.7% accuracy and outperforming contemporary methods across Accuracy, Precision, Recall, FPR, and FNR. Together, these contributions establish a progression from wavelet-based to curvelet-based and deep learning-driven SCI techniques, delivering robust, scalable, and highly precise solutions for forensic applications.

Source Camera Identification (SCI) is essential in digital image forensics, enabling reliable attribution of images to their originating devices for legal, investigative, and security applications. Yet, existing SCI methods often struggle under diverse imaging conditions due to scene-dependent noise and texture interference. This thesis advances SCI through four key contributions. First, a systematic evaluation of forty-two wavelets using the VISION dataset (34,427 images, 35 camera models, 11 brands) identified cdf9/7 as the most effective for Sensor Pattern Noise (SPN) extraction, followed by sym2 and coif1. Second, an Improved Camera Source Identification using Wavelet Noise Residuals and Texture Filtering (ICSI-WNRTF) method was developed to suppress high-texture regions, achieving 99% accuracy for model identification and 98% for device attribution. Third, a Curvelet-Based Camera Source Identification Leveraging Image Smooth Regions (CBCSI-SR) framework was introduced. By exploiting multi-scale directional features and isolating smooth regions, it achieved 99.6% model-level and 98.9% device-level accuracy while reducing false decisions. Finally, a Deep Learning-Based Texture Exclusion for Source Camera Identification (DLTESCI) approach combined texture suppression with a fine-tuned ResNet50, reaching 99.7% accuracy and outperforming contemporary methods across Accuracy, Precision, Recall, FPR, and FNR. Together, these contributions establish a progression from wavelet-based to curvelet-based and deep learning-driven SCI techniques, delivering robust, scalable, and highly precise solutions for forensic applications.