Gradient Contours Research Articles

Hierarchical Fusion of Infrared and Visible Images Based on Channel Attention Mechanism and Generative Adversarial Networks

In order to solve the problem that existing visible and infrared image fusion methods rely only on the original local or global information representation, which has the problem of edge blurring and non-protrusion of salient targets, this paper proposes a layered fusion method based on channel attention mechanism and improved Generative Adversarial Network (HFCA_GAN). Firstly, the infrared image and visible image are decomposed into a base layer and fine layer, respectively, by a guiding filter. Secondly, the visible light base layer is fused with the infrared image base layer by histogram mapping enhancement to improve the contour effect. Thirdly, the improved GAN algorithm is used to fuse the infrared and visible image refinement layer, and the depth transferable module and guided fusion network are added to enrich the detailed information of the fused image. Finally, the multilayer convolutional fusion network with channel attention mechanism is used to correlate the local information of the layered fusion image, and the final fusion image containing contour gradient information and useful details is obtained. TNO and RoadSence datasets are selected for training and testing. The results show that the proposed algorithm retains the global structure features of multilayer images and has obvious advantages in fusion performance, model generalization and computational efficiency.

Electro-osmotic transport and thermal energy dynamics of tetra-hybrid nano fluid in complex peristaltic flows

BackgroundThis study explores the use of tetra-hybrid nanoparticles consisting of gold, silver, alumina, and titania nanocomposites dispersed in a non-Newtonian Casson fluid modelled as blood. The motivation lies in harnessing the synergistic optical, electrical, thermal, and physicochemical properties of the multimodal nanoscale assembly to advance electrokinetic pumping processes. ObjectivesThe study aims to computationally investigate the complex transport assisted by tailored gold, silver, alumina, and titania dioxide nanoparticles in the tetra-hybrid nanofluid, with potential applications in targeted drug delivery, hyperthermia cancer treatment, Lab-on-Chip devices, and miniaturized biosensors. MethodologyThe constructed streaming flow model examines the cumulative influence of viscous heating, Joule heating, conductive thermal, and external laser irradiation. Current simulations examine Casson flows with the integrated nanoparticles by modelling two- and three-dimensional conduits subject to transverse magnetohydrodynamics and localized laser irradiation. Key findingsThe solutions provide mathematical predictions and physical insights into the fully coupled velocity, temperature, concentration, and pressure gradient contours. Additional plots examining dimensionless analytes against pertinent profiles, combined with streamlined visualization, further elucidate the multifaceted transport and complex trapping patterns stemming from non-Newtonian rheology. Applications/implicationsThis multiscale examination reveals performance improvements realizable by leveraging biocompatible tetra-hybrid nanocomposites in electrokinetic flows vital for next-generation biomedical technologies.

Effects of film cooling injection inclination angle on cooling performance in rotating detonation combustors

Film cooling is a promising thermal management solution for rotating detonation combustors (RDCs) maturing toward long-duration engineering implementation. Aimed at elucidating the interaction between air coolant and rotating detonation waves (RDWs) and assessing the cooling performance, three-dimensional numerical simulations are conducted on an RDC utilizing four different film cooling injection inclination angles and compared to a case without coolant injection. Increasing injection angles from 30° to 90° results in a broader detachment region and deeper penetration, negatively influencing the cooling performance. A time-averaged method is adopted to evaluate the overall cooling performance, including axial temperature profiles, film protection coverage, RDC film effectiveness, and pattern factor. The results show that the cylindrical cooling hole with a 30° injection angle outperforms others due to enhanced wall attachment of the coolant and reduced interaction with the mainstream hot gas. Consequently, a low injection angle within the manufacturing limits is recommended for practical applications. Furthermore, this study uncovers several phenomena unique to RDCs when introducing film cooling, absent in conventional gas turbines, such as temperature discrepancy between the inner and outer walls, elevated upstream temperature caused by coolant injection, and non-uniform cooling effectiveness between the two sides of the cooling holes. Finally, the interplay between film cooling and RDW is illustrated through temperature and pressure gradient contours.

Prototype Network for Predicting Occluded Picking Position Based on Lychee Phenotypic Features

The automated harvesting of clustered fruits relies on fast and accurate visual perception. However, the obscured stem diameters via leaf occlusion lack any discernible texture patterns. Nevertheless, our human visual system can often judge the position of harvesting points. Inspired by this, the aim of this paper is to address this issue by leveraging the morphology and the distribution of fruit contour gradient directions. Firstly, this paper proposes the calculation of fruit normal vectors using edge computation and gradient direction distribution. The research results demonstrate a significant mathematical relationship between the contour edge gradient and its inclination angle, but the experiments show that the standard error projected onto the Y-axis is smaller, which is evidently more conducive to distinguishing the gradient distribution. Secondly, for the front view of occluded lychee clusters, a fully convolutional, feature prototype-based one-stage instance segmentation network is proposed, named the lychee picking point prediction network (LP3Net). This network can achieve high accuracy and real-time instance segmentation, as well as for occluded and overlapping fruits. Finally, the experimental results show that the LP3Net based on this study, along with lychee phenotypic features, achieves an average location accuracy reaching 82%, significantly improving the precision of harvesting point localization for lychee clusters.

Optimization with manufacturing constraints for composite laminates reinforced by curvilinear fibers through a parametric level set method

This paper presents the design optimization with manufacturing constraints for composite laminates reinforced by curvilinear fibers by using a parametric level set method. In each layer of the composite laminate, a level set function is defined by using the compactly supported radial basis functions (CS-RBFs), and the iso-contours of each level set function represent the fiber paths in one layer of the laminate. By this means, the fiber orientation at an arbitrary point in a layer can be computed by the orientation of the tangent vector of the iso-contour passing through this point. To avoid structure defects (such as gap/overlap, wrinkling and delamination) that would arise if the fiber paths are not parallel or their curvature is too large, the gradient norm and contour curvature radius of the level set functions are considered as constraints in the design optimization. The p-norm method is used to aggregate the constraints to improve the optimization efficiency. The coefficients of CS-RBFs are considered as design variables. Several numerical examples of compliance minimization and eigen-frequency maximization are investigated to verify the proposed method.

Long-term borehole energy storage by the inlet position control for low temperature heat source application

Borehole thermal energy storage (BTES) systems are gaining attention due to their potential to efficiently store and release thermal energy. However, the conventional BTES systems often suffer from reduced efficiency under low heat source temperature conditions. This study aims to address this issue by introducing and evaluating the inlet position control method, a novel approach to enhance the BTES performance for low-temperature heat source applications. A single-cycle 3D simulation model, substantiated by experimental data from a BTES system installed in the YeonCheon region, Korea, is developed using the ANSYS Fluent program. The model investigates the heat transfer from the water heat source to the soil and compares the performance of the conventional method with the inlet position control method in terms of heat gradient contours and average temperature difference in the soil. The inlet position control method is found to improve the BTES storage efficiency by 11.90% during the early winter season of the thermal release period in the third year, compared to the conventional method. The standard deviation for simulation data is ± 1.787 with a peak error of 2.4%, indicating the effectiveness of the simulation model. Furthermore, the inlet control method enhances the BTES storage efficiency under low heat source temperature conditions and promotes faster thermal equalization, resulting in a 4.2% improvement in thermal energy release efficiency. We demonstrated the novelty of the inlet position control method in improving BTES performance and its applicability to low-temperature heat source applications. The findings have significant implications for the design and operation of future BTES systems, with the potential to expand their application in various industries and climates.

Giza pyramids construction algorithm with gradient contour approach for multilevel thresholding color image segmentation

Giza pyramids construction algorithm with gradient contour approach for multilevel thresholding color image segmentation

Flow characteristics and regime transition of aqueous foams in porous media over a wide range of quality, velocity, and surfactant concentration

Aqueous foam is broadly applicable to enhanced oil recovery (EOR). The rheology of foam as a function of foam quality, gas and liquid velocities, and surfactant concentration constitute the foundation of its application. The great variations of the above factors can affect the effectiveness of N2 foam in EOR continuously in complex formations, which is rarely involved in previous relevant studies. This paper presents an experimental study of foam flow in porous media by injecting pre-generated N2 foam into a sand pack under the conditions of considering a wide range of gas and liquid velocities and surfactant concentrations. The results show that in a wide range of gas and liquid velocities, the pressure gradient contours are L-shaped near the coordinate axes, but V-shaped in other regions. And the surfactant concentration is a strong factor influencing the trend of pressure gradient contours. Foam flow resistance is very sensitive to the surfactant concentration in both the high- and low-foam quality regime, especially when the surfactant concentration is less than CMC. The foam quality is an important variable to the flow resistance obtained. There exists a transition point from low- to high-quality regime in a particular flow system, where has the maximum flow resistance, the corresponding foam quality is called transition foam quality, which increases as the surfactant concentration increases. The results can add to our knowledge base of foam rheology in porous media, and can provide a strong basis for the field application of foams.

Application and improvement of Canny edge-detection algorithm for exterior wall hollowing detection using infrared thermal images

The periodic hollowing inspection of the existing building exterior wall is crucial for public safety and building energy conservation. Due to its non-destructive and intuition advantage, the infrared thermal detection is proposed to be an ideal survey method. However, much manual participation is required to distinguish the hollowing flaw relying on empirical judgment, and a heavy burden comes up when large-area diagnosis required. In order to improve the efficiency of hollowing detection, this investigation developed the Canny algorithm to realize the automatic processing using the computer instead of manual judgment.At first, reasonable pieces of setting advice were given to get more clear hollowing region contours with the final recognition outcome comparison of different processing methods for each step. Besides, it was found that the hollowing contour gradient values are lower and exist in a short interval, and the segmentation threshold value was critical in the Canny edge-detection algorithm, which highly restricted the speed of processing large amounts of infrared images. To improve the efficiency of thermal image recognition, a threshold selection method based on the local maximum inter-class variance algorithm was introduced into the Canny edge-detection algorithm. Compared with Sobel, Roberts, Prewitt, and LoG, the proposed algorithm presented a better performance in the identification of hollowing edge contour according to the verification based on three cases. It revealed that the improved Canny edge-detection algorithm was effective and efficient, which could not only eliminate the influence of subjective factors but also achieve full-automatic and batch processing.

Investigation of thermal entropy generation and nanofluid flow in a new heatsink with effect of nanoparticles shape

In this paper, the entropy generation (ENT) in two novel micro heatsinks (MHSs) is investigated numerically. The HSs are two types of octagonal ones with circular and rectangular fins. Water/alumina boehmite nanofluid (NF) is used as a cooling fluid for HSs, which enters from the middle of the heatsink (HS) and exits from four side outputs. The Reynolds number (Re) varies from 300 to 1700 and the concentration of nanoparticles (NPs) changes from 0 to 6%. The shapes of NPs include platelet brick, blade, cylinder, and Oblate Spheroid (OS). A heat flux due to the operation of a microprocessor is applied to the bottom of the HSs made of copper. The finite element method is used for the simulations. The variables include pumping power (PP), ENT, ENT due to fluid friction, thermal ENT, temperature gradient contours, etc. The results of this study show that less ENT occurs in the HS with rectangular fins and less PP occurs in the HS with circular ones. Increasing the Re reduces the amount of ENT and increases the PP. The results demonstrate that the addition of NPs with high volume percentages is not suitable in terms of ENT and PP.

Automatic Lineaments Extraction using the Line Algorithm in the Denguélé District (North West of Ivory Coast)

The aim of this work was to apply the LINE Algorithm (Segment Extraction Algorithm) on Landsat 8 images for automatic lineament extraction in the Denguélé district. The Landsat 8 images had previously been subjected to the technique of Principal Component Analysis (PCA). After that, we implemented the LINE algorithm. Indeed, the LINE algorithm uses the following six (6) parameters : RADI (Radius of the filter) for improving the quality of the input image, GTHR (Threshold of the contour gradient), LTHR (Threshold of the contour length), FTHR (Threshold of mounting error), ATHR (Angular difference threshold between two contours ) and DTHR (Distance chaining threshold to link two contours ) for lineament discrimination. Analysis of the principal components PCA 1, PCA2 and PCA3 of bands 1, 2, 3, 4, 5 and 7 of the Landsat 8 images shows that they contain respectively 79.57; 15.88 and 2.15%, this represents overall 97.6% of all channels. 3468 lineaments were extracted. The minimum and maximum lengths of the lineaments extracted are respectively 4201.08 m and 16167.59 m and their cumulative length is 18 919 517.9 m. The lineaments average lengths are 5.55 km; 5.75 km; 5.6 km and 5.40 km respectively for NE-SW, NS, E-W and NW-SE directions. The analysis of the directions of the lineaments using a rose diagram with 10 ° of frequency, shows that the dominant directions are NE-SW (31.83% of the total lineaments), EW (28.71% of the total lineaments) and NS (27.91% of the total lineaments).

Bio-Inspired Computing-A Dive into Critical Problems, Potential Architecture and Techniques

The integration and development of electronics in the recent years have impacted a major development on the world and humans, one among that is nanotechnology. Nanotechnology has achieved a greater progress in biomedical engineering in diagnosis and treatment, leading to the introduction of nanomaterials for drug delivery, prostheses and implanting. This work describes the Bio-Nano-tools that are developed based on iron oxide properties, automated tools used in the tumor detection, satin bowerbird optimization (SBO) technique employed in diagnosis of skin cancer. This work also highlights the post introduction development of nanomaterials like combination of nanotechnology with Artificial Intelligence (AI) and its impact, advancement of nanomaterials based on their operations, shapes and characteristics that leading to the growth of nanostructures with operations control properties. The paper also highlights the improvement of silicon neuromorphic photonic processors and parallel simulators in the development of bio inspired computing. We are hopeful that this review article provides future directions in Bio-Inspired Computing.
 HIGHLIGHTS
 
 In processing of medical images, noise plays a challenging role. So, reduction of noise is important, with the data that is analyzed in our review, it is shown that noise reduction can be achieved using Gradient and Feature Adaptive Contour (GFAC) model, with effective results
 There are many algorithms that are used for skin cancer detection, as highlighted in our review. Amongst all the methods, the particle swarm optimization (PSO) algorithm shows impressive results when compared to other models in terms of feature extraction in dermoscopy images
 Satin bowerbird optimization (SBO) algorithm helps in improving the CNN efficiency. The optimal justification of the hyper parameter numbers in convolutional neural network (CNN) for skin cancer diagnosis can be achieved using an SBO algorithm

Simulation of Foam Enhanced-Oil-Recovery Processes Using Operator-Based Linearization Approach

SummaryFoam injection is a promising enhanced-oil-recovery (EOR) technology that significantly improves the sweep efficiency of gas injection. Simulation of foam/oil displacement in reservoirs is an expensive process for conventional simulation because of the strongly nonlinear physics, such as multiphase flow and transport with oil/foam interactions. In this work, an operator-based linearization (OBL) approach, combined with the representation of foam by an implicit-texture (IT) model with two flow regimes, is extended for the simulation of the foam EOR process. The OBL approach improves the efficiency of the highly nonlinear foam-simulation problem by transforming the discretized nonlinear conservation equations into a quasilinear form using state-dependent operators. The state-dependent operators are approximated by discrete representation on a uniform mesh in parameter space. The numerical-simulation results are validated by using three-phase fractional-flow theory for foam/oil flow. Starting with an initial guess depending on the fitting of steady-state experimental data with oil, the OBL foam model is regressed to experimental observations using a gradient-optimization technique. A series of numerical validation studies is performed to investigate the accuracy of the proposed approach. The numerical model shows good agreement with analytical solutions at different conditions and with different foam parameters. With finer grids, the resolution of the simulation is better, but at the cost of more expensive computations. The foam-quality scan is accurately fitted to steady-state experimental data, except in the low-quality regime. In this regime, the used IT foam model cannot capture the upward-tilting pressure gradient (or apparent viscosity) contours. 1D and 3D simulation results clearly demonstrate two stages of foam propagation from inlet to outlet, as seen in the computed-tomography (CT) coreflood experiments: weak foam displaces most of the oil, followed by a propagation of stronger foam at lower oil saturation. OBL is a direct method to reduce nonlinearity in complex physical problems, which can significantly improve computational performance. Taking its accuracy and efficiency into account, the data-driven OBL-based approach could serve as a platform for efficient numerical upscaling to field-scale applications.

Automatic Impervious Surface Area Detection Using Image Texture Analysis and Neural Computing Models with Advanced Optimizers

Up-to-date information regarding impervious surface is valuable for urban planning and management. The objective of this study is to develop neural computing models used for automatic impervious surface area detection at a regional scale. To achieve this task, advanced optimizers of adaptive moment estimation (Adam), a variation of Adam called Adamax, Nesterov-accelerated adaptive moment estimation (Nadam), Adam with decoupled weight decay (AdamW), and a new exponential moving average variant (AMSGrad) are used to train the artificial neural network models employed for impervious surface detection. These advanced optimizers are benchmarked with the conventional gradient descent with momentum (GDM). Remotely sensed images collected from Sentinel-2 satellite for the study area of Da Nang city (Vietnam) are used to construct and verify the proposed approach. Moreover, texture descriptors including statistical measurements of color channels and binary gradient contour are employed to extract useful features for the neural computing model-based pattern recognition. Experimental result supported by statistical test points out that the Nadam optimizer-based neural computing model has achieved the most desired predictive accuracy for the data collected in the studied region with classification accuracy rate of 97.331%, precision = 0.961, recall = 0.984, negative predictive value = 0.985, and F1 score = 0.972. Therefore, the model developed in this study can be a helpful tool for decision-makers in the task of urban land-use planning and management.

Estimation of Geothermal Gradient, Geothermal Heat Flux and Thermal Conductivity of Rocks in Western Niger Delta Using Well Log Data

This study presents the result of the estimation of heat flow from six (6) wells (Well X:001 to 006) in South-Western Niger Delta using values of Geothermal gradient (GG), Geothermal heat flux (Q) and thermal conductivity (K) computed from Sonic and continuous temperature log data for each well. Geothermal gradient was computed from continuous temperature logs using the simple gradient method while geothermal heat flux and thermal conductivity of the rocks in the wells were computed from the sonic log data, using the Relative Heat Flow Model and Fourier One-dimensional Heat Flow Law respectively. The results were analysed and interpreted to investigate the thermal structure and pattern of heat flow distribution of the basin. Results showed that geothermal gradient ranges from 1.45 0 C/100m to a value of 1.61 0 C/100m, with a simple average of 1.55 0 C/100m. Geothermal gradient contour map computed from this result, showed a low thermal gradient at the northern part of the study area where we have Well X-006 and increases outwards in all direction as we move further offshore. These differences reflect changes in thermal conductivity of rocks, ground water movement and endothermic reaction during diagenesis, since geothermal gradient is influenced by lithology or differential rate of sedimentation. Therefore, it was inferred that sediments with a relatively high geothermal gradient (1.55 to 1.61 0 C/100m) will mature earlier (low oil window) than those with low thermal gradient values. By implication, a high geothermal gradient enhances the early formation of oil at relatively shallow burial depths, but causes the depth range of the oil window to be narrow, while low geothermal gradient causes the first formation of oil to begin at fairly deep subsurface levels, but makes the oil window broad. Geothermal heat flux estimated from subsurface temperature and one-way sound travel time, shows heat flux varying between 33.16 mWm -2 to 72.73 mWm -2 with a simple average of 48.43 mWm -2 . Low heat flux was observed at the central part of the study area which increases towards the western and eastern parts of the area with Well X-005 characterized by a higher geothermal heat flux. Therefore, it was inferred that the western and eastern parts of the study area with higher heat flux values may be characterized as zones with maximum sediment thickness and are characterized as having depressions (gravity low) on the geoid which is characteristics of a basin, while the central part of the study area with low heat flux values correspond with zones of minimum sediment thickness. Also, thermal conductivity of rocks in the study area computed directly from heat flux and geothermal gradient results, ranges from 2.28W/m 0 C to 4.76 W/m 0 C with an average of 3.19 W/m 0 C. Thermal conductivity contour map computed from this result, showed low thermal conductivity values observed at the central part of the study area, and increases outwards towards the west and eastern parts. This pattern of thermal conductivity variation suggests probably there exists heavy crude oil at the central part of the study area and lighter crude oil as we move outward in all direction. It was also observed that within each well, thermal conductivity increased with depth and decreased with porosity which may be caused by difference in lithology and fluid content, due to the fact that all pore fillers (i:e gases and liquids) are poor conductors. The estimated values of geothermal gradient, heat flux and thermal conductivity obtained in this study are similar to the results obtained from previous studies in the region and with other passive continental margins of the world. Keywords: Heat flow, Geothermal Gradient, Geothermal Heat Flux, Thermal Conductivity, Continuous Temperature and Sonic log. DOI: 10.7176/JETP/10-2-04 Publication date: April 30 th 2020

Adaptive iterative attack towards explainable adversarial robustness

Image classifiers based on deep neural networks show severe vulnerability when facing adversarial examples crafted on purpose. Designing more effective and efficient adversarial attacks is attracting considerable interest due to its potential contribution to interpretability of deep learning and validation of neural networks’ robustness. However, current iterative attacks use a fixed step size for each noise-adding step, making further investigation into the effect of variable step size on model robustness ripe for exploration. We prove that if the upper bound of noise added to the original image is fixed, the attack effect can be improved if the step size is positively correlated with the gradient obtained at each step by querying the target model. In this paper, we propose Ada-FGSM (Adaptive FGSM), a new iterative attack that adaptively allocates step size of noises according to gradient information at each step. Improvement of success rate and accuracy decrease measured on ImageNet with multiple models emphasizes the validity of our method. We analyze the process of iterative attack by visualizing their trajectory and gradient contour, and further explain the vulnerability of deep neural networks to variable step size adversarial examples.

Image Processing‐Based Spall Object Detection Using Gabor Filter, Texture Analysis, and Adaptive Moment Estimation (Adam) Optimized Logistic Regression Models

This study aims at proposing a computer vision model for automatic recognition of localized spall objects appearing on surfaces of reinforced concrete elements. The new model is an integration of image processing techniques and machine learning approaches. The Gabor filter supported by principal component analysis and k‐means clustering is used for identifying the region of interest within an image sample. The binary gradient contour, gray level co‐occurrence matrix, and color channels’ statistical measurements are employed to compute the texture of the extracted region of interest. Based on the computed texture‐based features, the logistic regression model trained by the state‐of‐the‐art adaptive moment estimation (Adam) is utilized to establish a decision boundary that delivers predictions on the status of “nonlocalized spall” and “localized spall.” Experimental results demonstrate that the newly developed model is able to achieve good detection accuracy with classification accuracy rate = 85.32%, precision = 0.86, recall = 0.79, negative predictive value = 0.85, and F1 score = 0.82. Thus, the proposed computer vision model can be helpful to assist decision makers in the task of the periodic survey of structure heath condition.

Computer Algorithm for Archaeological Projectile Points Automatic Classification

The manual archaeological projectile point morphological classification is an extensive and complex process since it involves a large number of categories. This article presents an algorithm that automatically makes this process, based on the projectile point digital image and using a classification scheme according to global archaeological approaches. The algorithm supports different conditions such as changes in scale and quality of the image. Moreover, it requires only a uniform background and an approximate north--south projectile point orientation. The principal computer methods that compose the algorithm are the curvature scale space map (CSS-map), the gradient contour on the projectile point, and the support vector machines (SVM) algorithm. Finally, the classifier was trained and tested on a dataset of approximately 800 projectile points images, and the results have shown a better performance than other shape descriptors such as Pyramid of Histograms of Orientation Gradients (PHOG), Histogram of Orientation Shape Context (HOOSC) (both used in a bag-of-words context), and geometric moment invariants (Hu moments).

Early Detection of Skin Cancer Using Melanoma Segmentationtechnique.

The significance of pattern recognition techniques is widely enhanced in image processing and medical applications. Thus, lesion segmentation method is an essential technique of pattern recognition algorithms to detect the melanoma skin cancer in patients at earliest stage, otherwise, in further stages it becomes one of the deadliest disease and its mortality rate is very high. Therefore, a precise melanoma segmentation technique is introduced based on the Gradient and Feature Adaptive Contour (GFAC) model to detect melanoma skin cancer in earliest stage and diagnosis of dermoscopic images. In the proposed image segmentation technique pre-processing and noise elimination techniques are introduced to decrease noise and make execution faster. This technique helps in separating the required entity from the background and gather the information from the adjacent pixels of similar classes. Multiple Gaussian distributed patterns are adopted to extract efficient features and to get precise segmentation. The proposed GFACmodel is noise free and consist of smoother border. The segmentation model efficiency is tested on PH2 dataset. The superiority of the proposed modified gradient and feature adaptive contour model can be verified against various state-of-art-techniques in terms of segmented image, error reduction and efficient feature extraction.

Segmentation of cervical nuclei using SLIC and pairwise regional contrast.

A framework to detect and segment nuclei from cervical cytology images is proposed in this study. Poor contrast, spurious edges, degree of overlap, and intensity inhomogeneity make the nuclei segmentation task more complex in overlapping cell images. The proposed technique segments cervical nuclei by merging over-segmented SLIC superpixel regions using a novel region merging criteria based on pairwise regional contrast and image gradient contour evaluations. The framework was evaluated using the first overlapping cervical cytology image segmentation challenge - ISBI 2014 dataset. The result shows that the proposed framework outperforms the state-of-the-art algorithms in nucleus detection and segmentation accuracies.