Image Texture Research Articles

Recovering NDVI over lake surfaces: Initial insights from CYGNSS data enhanced by ERA-5 inputs

The escalating water pollution in many lakes has led to more frequent occurrences of algal bloom disasters in recent decades. The severity of these disasters can be assessed through remote sensing techniques, specifically using the Normalized Difference Vegetation Index (NDVI) for measurement. However, NDVI observations using optical sensors are often affected by cloud and fog in areas with numerous water bodies, such as Taihu Lake. Sensors operating in the microwave band can effectively mitigate this issue, particularly the emerging Global Navigation Satellite System Reflectometry (GNSS-R), which offers high temporal resolution and cost-effectiveness. In this paper, we propose a new method to recover lake-surface NDVI on cloudy days, utilizing GNSS-R observables and auxiliary meteorological data in conjunction with a machine learning regression algorithm called Bagging Tree. We also examine the effective range of GNSS-R data within this application scenario. Meanwhile, the Weighted Linear Regression-Laplacian Prior Regulation Method (WLR-LPRM) image gap-filling algorithm is used as a benchmark to evaluate recovery accuracy. The regression coefficient of NDVI retrieved using the proposed method is 0.95, with a root mean square error (RMSE) of 0.021 and a mean absolute error (MAE) of 0.010. Compared to the previous work on GNSS-R algal bloom detection with overall accuracy of 0.82, this work shows significant improvement in both accuracy and utility. The recovery of lake surface NDVI provides detailed insights into algal blooms, including quantifiable metrics such as the amount and spatial distribution, which are crucial for effective monitoring and management. Additionally, the recovered image textures exhibit high clarity and closely resemble the reference NDVI images. Experimental evaluation using simulated and actual cloud blocks indicates the model’s robustness to recover NDVI under varying cloud cover conditions. In summary, this study demonstrates the capability of GNSS-R aided by supplementary data for recovering missing NDVI values on lake surfaces when optical observations are absent for the first time.

Effect of masticatory behavior on the characteristics of biscuit bolus with different amounts of chia seed powder without compromising taste

This study investigated the impact of masticatory behavior, including variations in chewing duration (D) and chewing frequency (F), on the bolus characteristics of biscuits with different chia seed powder (CSP) additions. Five distinct biscuits were prepared with varying amounts of CSP (0%, 5%, 10%, 15%, and 20%, w/w) and their physicochemical indexes and sensory properties were conducted. Subsequently, six types of bolus were obtained under different conditions of chewing duration (15, 30, or 45s) and chewing frequencies (70 or 90 times per minute). The moisture content, texture properties, and in vitro starch digestion of bolus were also determined. The findings indicated a significant decrease in the spread ratio, with a reduction of approximately 25% in biscuits enriched with 20% (w/w) CSP compared to the control (0%). And the 15% added CSP biscuits had the highest overall acceptance rating. After chewing, the particle size decreased significantly (p<0.05), with increased surface inhomogeneity and a decline in rapidly digestible starch (RDS) levels. The moisture content, salivary uptake, and resistant starch (RS) of bolus chewed for a long duration (D45) were at least 1.3 times higher than those chewed for a short duration (D15). Nevertheless, the salivary flow rate and RDS were found to be significantly lower. Changes in chewing frequency for the bolus with CSP did not significantly alter the bolus texture and image texture properties. However, the discomfort associated with high-frequency reduced the saliva flow rate. This study discusses how CSP addition and masticatory behavior affect the characteristics of CSP biscuit bolus, providing insights for innovative chia seed nutritious food research and understanding consumer masticatory behavior.

Longitudinal FDG-PET Radiomics for Early Prediction of Treatment Response to Chemoradiation in Locally Advanced Cervical Cancer: A Pilot Study

Objectives: This study aimed to assess the capacity of longitudinal FDG-PET radiomics for early distinguishing between locally advanced cervical cancer (LACC) patients who responded to treatment and those who did not. Methods: FDG-PET scans were obtained before and midway through concurrent chemoradiation for a study cohort of patients with LACC. Radiomics features related to image textures were extracted from the primary tumor volumes and stratified for relevance to treatment response status with the aid of random forest recursive feature elimination. Predictive models based on the k-nearest neighbors time series classifier were developed using the top-selected features to differentiate between responders and non-responders. The performance of the developed models was evaluated using receiver operating characteristic (ROC) curve analysis and n-fold cross-validation. Results: The top radiomics features extracted from scans taken midway through treatment showed significant differences between the two responder groups (p-values &lt; 0.0005). In contrast, those from pretreatment scans did not exhibit significant differences. The AUC of the mean ROC curve for the predictive model based on the top features from pretreatment scans was 0.8529, while it reached 0.9420 for those derived midway through treatment scans. Conclusions: The study highlights the potential of longitudinal FDG-PET radiomics extracted midway through treatment for predicting response to chemoradiation in LACC patients and emphasizes that interim PET scans could be crucial in personalized medicine, ultimately enhancing therapeutic outcomes for LACC.

An automated high-content screening and assay platform for the analysis of spheroids at subcellular resolution.

The endomembrane system is essential for healthy cell function, with the various compartments carrying out a large number of specific biochemical reactions. To date, almost all of our understanding of the endomembrane system has come from the study of cultured cells growing as monolayers. However, monolayer-grown cells only poorly represent the environment encountered by cells in the human body. As a first step to address this disparity, we have developed a platform that allows us to investigate and quantify changes to the endomembrane system in three-dimensional (3D) cell models, in an automated and highly systematic manner. HeLa Kyoto cells were grown on custom-designed micropatterned 96-well plates to facilitate spheroid assembly in the form of highly uniform arrays. Fully automated high-content confocal imaging and analysis were then carried out, allowing us to measure various spheroid-, cellular- and subcellular-level parameters relating to size and morphology. Using two drugs known to perturb endomembrane function, we demonstrate that cell-based assays can be carried out in these spheroids, and that changes to the Golgi apparatus and endosomes can be quantified from individual cells within the spheroids. We also show that image texture measurements are useful tools to discriminate cellular phenotypes. The automated platform that we show here has the potential to be scaled up, thereby allowing large-scale robust screening to be carried out in 3D cell models.

AFDFusion: An adaptive frequency decoupling fusion network for multi-modality image

The multi-modality image fusion goal is to create a single image that provides a comprehensive scene description and conforms to visual perception by integrating complementary information about the merits of the different modalities, e.g., salient intensities of infrared images and detail textures of visible images. Although some works explore decoupled representations of multi-modality images, they struggle with complex nonlinear relationships, fine modal decoupling, and noise handling. To cope with this issue, we propose an adaptive frequency decoupling module to perceive the associative invariant and inherent specific among cross-modality by dynamically adjusting the learnable low frequency weight of the kernel. Specifically, we utilize a contrastive learning loss for restricting the solution space of feature decoupling to learn representations of both the invariant and specific in the multi-modality images. The underlying idea is that: in decoupling, low frequency features, which are similar in the representation space, should be pulled closer to each other, signifying the associative invariant, while high frequencies are pushed farther away, also indicating the intrinsic specific. Additionally, a multi-stage training manner is introduced into our framework to achieve decoupling and fusion. Stage I, MixEncoder and MixDecoder with the same architecture but different parameters are trained to perform decoupling and reconstruction supervised by the contrastive self-supervised mechanism. Stage II, two feature fusion modules are added to integrate the invariant and specific features and output the fused image. Extensive experiments demonstrated the proposed method superiority over the state-of-the-art methods in both qualitative and quantitative evaluation on two multi-modal image fusion tasks.

Dual-image reversible data hiding based on pixel value parity and multiple embedding strategy

Dual-image reversible data hiding (DI-RDH) has attracted a lot of attention for its excellent embedding capability. However, the stego image after data embedding is degraded, which limits its practical application. This paper proposes an innovative DI-RDH method based on pixel value parity (PVP), which prevents influences from the image texture features and ensures that the pixel value deviation between the temporary stego image and the original image is −1, 0, or 1, as it can partially offset the difference introduced when generating the cover image. Such characteristics enable us to establish a multiple embedding strategy (MES), which applies the difference image to embed the secret message, decreasing the number of invalid shifting pixels in the difference histogram to enhance embedding capacity and image quality. The average maximum embedding rate of our method combining PVP and two phases of MES is determined to be 1.37 bpp, corresponding to the theoretical results. The average peak signal-to-noise ratio of our method on the UCID database is increased by at least 2.19 dB for a given ER of 0.5 bpp compared with several state-of-the-art methods.

Recognition of Heat-Damaged Corn Seeds Based on Fusion of Laser Ultrasonic Signal and Infrared Image Features

Corn is widely cultivated on a global scale. However, high temperatures during storage and transportation can lead to thermal damage to the kernels, negatively impacting their quality. Traditional methods for identifying heat-damaged grains primarily rely on manual inspection, which is characterized by low efficiency and accuracy. This study proposes a novel identification method that integrates laser ultrasonic signals with infrared image texture features. A pulsed laser stimulates the seeds to generate laser ultrasonic signals, while an infrared camera captures infrared images of the seeds. We extract time-domain, frequency-domain, and Hilbert-domain features from the laser ultrasonic signals, in addition to texture features from the infrared images. These features are combined using Canonical Correlation Analysis (CCA). Subsequently, the fused features are classified using a Backpropagation (BP) neural network, Support Vector Machine (SVM), and Particle Swarm Optimization–Support Vector Machine (PSO–SVM). The results indicate that the recognition rate achieved with the fused ‘signal-image’ features reaches 99.17%, providing a novel approach for detecting heat-damaged corn seeds.

Tree Species Classification by Multi-Season Collected UAV Imagery in a Mixed Cool-Temperate Mountain Forest

Effective forest management necessitates spatially explicit information about tree species composition. This information supports the safeguarding of native species, sustainable timber harvesting practices, precise mapping of wildlife habitats, and identification of invasive species. Tree species identification and geo-location by machine learning classification of UAV aerial imagery offer an alternative to tedious ground surveys. However, the timing (season) of the aerial surveys, input variables considered for classification, and the model type affect the classification accuracy. This work evaluates how the seasons and input variables considered in the species classification model affect the accuracy of species classification in a temperate broadleaf and mixed forest. Among the considered models, a Random Forest (RF) classifier demonstrated the highest performance, attaining an overall accuracy of 83.98% and a kappa coefficient of 0.80. Simultaneously using input data from summer, winter, autumn, and spring seasons improved tree species classification accuracy by 14–18% from classifications made using only single-season input data. Models that included vegetation indices, image texture, and elevation data obtained the highest accuracy. These results strengthen the case for using multi-seasonal data for species classification in temperate broadleaf and mixed forests since seasonal differences in the characteristics of species (e.g., leaf color, canopy structure) improve the ability to discern species.

Techniques for using backlighting to enhance the expressiveness of photographs

This article examines the historical and contemporary aspects of backlighting in photography, focusing on its visual and psychological impact and the technical and color features that enhance its effectiveness. Drawing on principles established during the Baroque era by artists like Caravaggio and Rembrandt, the study explores how modern photographers integrate these techniques with advanced lighting technologies and color theories, particularly those of Johannes Itten. Through a comprehensive analysis of backlighting applications in portrait and food photography, the research highlights how this technique enhances depth, texture, and emotional resonance in images. The findings demonstrate that mastery of backlighting, combined with technical proficiency and an understanding of color interactions, allows photographers to create compelling and emotionally impactful works. The novelty of this work lies in its synthesis of historical practices with modern technological advancements, offering new insights into the creative possibilities of backlighting in professional photography.

Wide-field scanning ghost imaging based on a local binary pattern and untrained neural network

Continuous scene imaging is an important research goal in the field of autonomous driving, and the key is to ensure the imaging quality and efficiency. In this paper, we propose a method for information fusion in wide-field scanning ghost imaging using a local binary pattern (LBP) based on deep learning. The initial physical model formed by the LBP integrated into a deep neural network, which effectively enhances the expression of image texture details. Then the collected bucket signals are used as labels for adaptive image reconstruction, enabling the acquisition of images at each scanning position without the need for training on any dataset. Moreover, by employing weighted fusion to combine the image data from each scanning position, which effectively eliminates gaps that arise from direct stitching. Both simulation and experimental results demonstrate that our approach is capable of achieving high-quality detailed imaging with fewer measurements. Additionally, we analyze the impact of the projection beam step length, finding that our method yields significantly better imaging quality with larger steps compared to other methods using smaller steps. Our research also has the application prospect in medical detection, remote sensing and other fields.

A Feature-Driven Inception Dilated Network for Infrared Image Super-Resolution Reconstruction

Image super-resolution (SR) algorithms based on deep learning yield good visual performances on visible images. Due to the blurred edges and low contrast of infrared (IR) images, methods transferred directly from visible images to IR images have a poor performance and ignore the demands of downstream detection tasks. Therefore, an Inception Dilated Super-Resolution (IDSR) network with multiple branches is proposed. A dilated convolutional branch captures high-frequency information to reconstruct edge details, while a non-local operation branch captures long-range dependencies between any two positions to maintain the global structure. Furthermore, deformable convolution is utilized to fuse features extracted from different branches, enabling adaptation to targets of various shapes. To enhance the detection performance of low-resolution (LR) images, we crop the images into patches based on target labels before feeding them to the network. This allows the network to focus on learning the reconstruction of the target areas only, reducing the interference of background areas in the target areas’ reconstruction. Additionally, a feature-driven module is cascaded at the end of the IDSR network to guide the high-resolution (HR) image reconstruction with feature prior information from a detection backbone. This method has been tested on the FLIR Thermal Dataset and the M3FD Dataset and compared with five mainstream SR algorithms. The final results demonstrate that our method effectively maintains image texture details. More importantly, our method achieves 80.55% mAP, outperforming other methods on FLIR Dataset detection accuracy, and with 74.7% mAP outperforms other methods on M3FD Dataset detection accuracy.

Enhancing Robustness and Imperceptibility with a Texture-Based Adaptive QIM Approach

This paper proposes a new approach for image data hiding that aims to improve both robustness and imperceptibility. The proposed approach is based on a texture-adaptive quantization index modulation (QIM) method that takes into account the local characteristics of the image texture to optimize the embedding process. The approach is evaluated using various image datasets and compared to existing state-of-the-art techniques. The results show that the proposed approach achieves better performance in terms of robustness against various image processing attacks, with BERs of zero in some attacks, while maintaining a high level of imperceptibility with PSNRs exceeding 46 decibels. It can be deduced that adaptation and normalization of the embedding strength arguments can enhance the QIM methods’ performance. The proposed approach has potential applications in image authentication, copyright protection, and data hiding.

Cardiac amyloidosis: a texture issue. a proof of concept study on quantification of granular sparkling in patients with transthyretin-related cardiac amyloidosis using radiomics and granulometry

Abstract Background Granular sparkling is a well-known echocardiographic feature found in patients with transthyretin-related cardiac amyloidosis (ATTR-CA). However, there is no objective technique for quantifying this feature, which therefore remains a qualitative, elusive and ultimately unreliable imaging characteristic. Recent evidence suggests that radiomics and, in particular, ultrasonomics could play an important role in the non-invasive tissue characterization of cardiomyopathies through the study of specific myocardial texture features. Purpose The aim of this study is to statistically and geometrically characterize granular sparkling as a volume-independent texture property of the myocardium in patients with ATTR-CA. Methods We retrospectively collected echocardiogram video-clips in parasternal long axis (PLAX) and 4-chamber (4CH) views of 229 patients with ATTR-CA, and 224 age- and gender-matched hypertensive patients without any known cardiac disease. From each video-clip, one end-diastole frame was extracted and annotated by an expert to identify a region-of-interest (ROI) within the interventricular septum. Left ventricle chamber masks were also extracted, and used as a brightness reference to enforce invariance w.r.t. the settings of the specific ultrasound system (Fig. 1). Because many established radiomic textural features are heavily volume-confounded, we analyzed the ROI texture by extracting a subset of volume-invariant radiomic features (i.e. ROI-independent), as well as morphological granulometry features. We then fitted a logistic regression classifier to discriminate between ATTR-CA and controls based on the computed textural features. Results 94 radiomic features were identified, of which 73 were volume-invariant. The texture-based classifier predicted the diagnosis with a cross-validated accuracy of 91%, a specificity of 90% and a sensitivity of 92% by utilizing 73 volume-invariant radiomic features on both PLAX and 4CH views (Fig. 2). The addition of granulometry and the further addition of 11 volume-confounded features and 9 shape features to the model did not significantly enhance its discriminative performance (Fig. 2). The top 10 volume-invariant discriminative features were largely consistent in PLAX and 4CH views (Fig. 2), with variables descriptive of greater heterogeneity (such as glszm_ZoneEntropy and glszm_RunEnNtropy) and a higher spatial rate of change (such as ngtdm_Complexity) in the texture of ATTR-CA frames as compared to control frames. Conclusions Our results confirm the ability of radiomics in detecting subtle differences in the myocardial texture of echocardiographic images between patients with ATTR-CA and hypertensive controls. High levels of accuracy are obtained with a purely texture-based, volume-independent set of features. These findings suggest that the diagnosis of ATTR-CA could be simplified and made earlier using a statistical model based on the extraction of radiomic features.Figure 1Figure 2

Comparison of image quality and lesion conspicuity between conventional and deep learning reconstruction in gadoxetic acid-enhanced liver MRI

ObjectiveTo compare the image quality and lesion conspicuity of conventional vs deep learning (DL)-based reconstructed three-dimensional T1-weighted images in gadoxetic acid-enhanced liver magnetic resonance imaging (MRI).MethodsThis prospective study (NCT05182099) enrolled participants scheduled for gadoxetic acid-enhanced liver MRI due to suspected focal liver lesions (FLLs) who provided signed informed consent. A liver MRI was conducted using a 3-T scanner. T1-weighted images were reconstructed using both conventional and DL-based (AIRTM Recon DL 3D) reconstruction algorithms. Three radiologists independently reviewed the image quality and lesion conspicuity on a 5-point scale.ResultsFifty participants (male = 36, mean age 62 ± 11 years) were included for image analysis. The DL-based reconstruction showed significantly higher image quality than conventional images in all phases (3.71–4.40 vs 3.37–3.99, p < 0.001 for all), as well as significantly less noise and ringing artifacts than conventional images (p < 0.05 for all), while also showing significantly altered image texture (p < 0.001 for all). Lesion conspicuity was significantly higher in DL-reconstructed images than in conventional images in the arterial phase (2.15 [95% confidence interval: 1.78, 2.52] vs 2.03 [1.65, 2.40], p = 0.036), but no significant difference was observed in the portal venous phase and hepatobiliary phase (p > 0.05 for all). There was no significant difference in the figure-of-merit (0.728 in DL vs 0.709 in conventional image, p = 0.474).ConclusionDL reconstruction provided higher-quality three-dimensional T1-weighted imaging than conventional reconstruction in gadoxetic acid-enhanced liver MRI.Critical relevance statementDL reconstruction of 3D T1-weighted images improves image quality and arterial phase lesion conspicuity in gadoxetic acid-enhanced liver MRI compared to conventional reconstruction.Key PointsDL reconstruction is feasible for 3D T1-weighted images across different spatial resolutions and phases.DL reconstruction showed superior image quality with reduced noise and ringing artifacts.Hepatic anatomic structures were more conspicuous on DL-reconstructed images.Graphical

Non-Destructive Monitoring of Sweet Pepper Samples After Selected Periods of Lacto-Fermentation

Fermented food is characterized by positive health-promoting properties. The objective of this study was to distinguish and assess the changes in the flesh structure of sweet bell pepper samples after specific periods of fermentation in a non-destructive manner. Two cultivars of pepper, red and yellow, were subjected to lacto-fermentation. The experiments lasted 56 days and the samples were taken for analysis at the beginning of the study (0 days) and after 3, 7, 10, 14, 21, 28, and 56 days. The fermentation process was monitored based on image features, which were used to develop machine learning models distinguishing samples before and after various periods of lacto-fermentation (0, 3, 7, 10, 14, 21, 28, and 56 days). The average accuracy of the classification of red bell pepper samples was up to 93% for the model built using IBk (Lazy group). The yellow bell pepper samples were distinguished up to 90% accuracy by the LMT algorithm (Trees group). The performed study allowed us to determine the changes in pepper flesh in terms of image textures during lacto-fermentation.

Research on Target Image Classification in Low-Light Night Vision.

In extremely dark conditions, low-light imaging may offer spectators a rich visual experience, which is important for both military and civic applications. However, the images taken in ultra-micro light environments usually have inherent defects such as extremely low brightness and contrast, a high noise level, and serious loss of scene details and colors, which leads to great challenges in the research of low-light image and object detection and classification. The low-light night vision image used as the study object in this work has an excessively dim overall picture and very little information about the screen's features. Three algorithms, HE, AHE, and CLAHE, were used to enhance and highlight the image. The effectiveness of these image enhancement methods is evaluated using metrics such as the peak signal-to-noise ratio and mean square error, and CLAHE was selected after comparison. The target image includes vehicles, people, license plates, and objects. The gray-level co-occurrence matrix (GLCM) was used to extract the texture features of the enhanced images, and the extracted image texture features were used as input to construct a backpropagation (BP) neural network classification model. Then, low-light image classification models were developed based on VGG16 and ResNet50 convolutional neural networks combined with low-light image enhancement algorithms. The experimental results show that the overall classification accuracy of the VGG16 convolutional neural network model is 92.1%. Compared with the BP and ResNet50 neural network models, the classification accuracy was increased by 4.5% and 2.3%, respectively, demonstrating its effectiveness in classifying low-light night vision targets.

Need of Feature Extraction in Coconut Tree Disease Detection: A Review

Agriculture has served as the primary means of sustenance for mankind for thousands of years. Even in the present day, it continues to support approximately 50% of the global population. Plant diseases pose a significant threat to crop production, resulting in substantial losses each year on a global scale. To mitigate the financial impact of these diseases, it is crucial to maintain the health of plants throughout their growth and development. The symptoms of infections are predominantly visible on plant leaves, making them a common indicator for disease detection and identification. However, visually observing and identifying diseases is a challenging task that requires extensive human expertise. To provide farmers with improved assistance in disease detection, image processing techniques combined with computational intelligence or soft computing techniques can be employed. These methods offer a more effective approach to disease detection. By extracting features from the symptoms of a disease, it becomes possible to identify and detect the presence of a disease in plants. Therefore, feature extraction techniques play a vital role in such systems. This paper focuses on reviewing the feature extraction methods, highlighting their advantages and disadvantages. It provides a comprehensive discussion on various image features, including color, texture, and shape, for different types of disorders found in diverse agricultural practices.

Optimal Illumination Distance Metrics for Person Re-identification in Complex Lighting Conditions

Person Re-identification is extensively applied in public security and surveillance. However, environmental factors like time and location often lead to varying lighting conditions in captured pedestrian images, significantly impacting identification accuracy. Current approaches mitigate this issue through lighting transformation techniques, aiming to normalize images to a standard lighting condition for consistent person re-identification results. Yet, these methods overlook the fact that different content may hold distinct identification values under diverse lighting conditions. To address this, we conducted an analysis on the identification distance between images of the same or different pedestrians under predefined lighting conditions. From this analysis, we introduce the concept of optimal lighting: a condition where the distance between image pairs is minimized compared to other lighting scenarios. We propose utilizing this optimal lighting distance in the image retrieval process for final ranking. Our study, validated on synthetic datasets Market-IA and Duke-IA, demonstrates that optimal lighting is independent of image texture information. Each image pair exhibits a unique optimal lighting, yet consistently shows a minimum distance value.

Remote Sensing Image Denoising Based on Feature Interaction Complementary Learning

Optical remote sensing images are of considerable significance in a plethora of applications, including feature recognition and scene semantic segmentation. However, the quality of remote sensing images is compromised by the influence of various types of noise, which has a detrimental impact on their practical applications in the aforementioned fields. Furthermore, the intricate texture characteristics inherent to remote sensing images present a significant hurdle in the removal of noise and the restoration of image texture details. In order to address these challenges, we propose a feature interaction complementary learning (FICL) strategy for remote sensing image denoising. In practical terms, the network is comprised of four main components: noise predictor (NP), reconstructed image predictor (RIP), feature interaction module (FIM), and fusion module. The combination of these modules serves to not only complete the fusion of the prediction results of NP and RIP, but also to achieve a deep coupling of the characteristics of the two predictors. Consequently, the advantages of noise prediction and reconstructed image prediction can be combined, thereby enhancing the denoising capability of the model. Furthermore, comprehensive experimentation on both synthetic Gaussian noise datasets and real-world denoising datasets has demonstrated that FICL has achieved favorable outcomes, emphasizing the efficacy and robustness of the proposed framework.

Advanced Deep Learning Fusion Model for Early Multi-Classification of Lung and Colon Cancer Using Histopathological Images.

In recent years, the healthcare field has experienced significant advancements. New diagnostic techniques, treatments, and insights into the causes of various diseases have emerged. Despite these progressions, cancer remains a major concern. It is a widespread illness affecting individuals of all ages and leads to one out of every six deaths. Lung and colon cancer alone account for nearly two million fatalities. Though it is rare for lung and colon cancers to co-occur, the spread of cancer cells between these two areas-known as metastasis-is notably high. Early detection of cancer greatly increases survival rates. Currently, histopathological image (HI) diagnosis and appropriate treatment are key methods for reducing cancer mortality and enhancing survival rates. Digital image processing (DIP) and deep learning (DL) algorithms can be employed to analyze the HIs of five different types of lung and colon tissues. Therefore, this paper proposes a refined DL model that integrates feature fusion for the multi-classification of lung and colon cancers. The proposed model incorporates three DL architectures: ResNet-101V2, NASNetMobile, and EfficientNet-B0. Each model has limitations concerning variations in the shape and texture of input images. To address this, the proposed model utilizes a concatenate layer to merge the pre-trained individual feature vectors from ResNet-101V2, NASNetMobile, and EfficientNet-B0 into a single feature vector, which is then fine-tuned. As a result, the proposed DL model achieves high success in multi-classification by leveraging the strengths of all three models to enhance overall accuracy. This model aims to assist pathologists in the early detection of lung and colon cancer with reduced effort, time, and cost. The proposed DL model was evaluated using the LC25000 dataset, which contains colon and lung HIs. The dataset was pre-processed using resizing and normalization techniques. The model was tested and compared with recent DL models, achieving impressive results: 99.8% for precision, 99.8% for recall, 99.8% for F1-score, 99.96% for specificity, and 99.94% for accuracy. Thus, the proposed DL model demonstrates exceptional performance across all classification categories.