Image Vector Research Articles

A robust encryption watermarking algorithm for medical images based on ridgelet-DCT and THM double chaos

With the help of big data, cloud computing, artificial intelligence and other technologies, the informatization and intelligence of the wisdom medical have been gradually realized. However, with the transmission and storage of massive amounts of medical images in the cloud, information security issues have become increasingly prominent. The privacy of patients is at risk of disclosure, theft and tampering, which has become an important challenge restricting the development of wisdom medical. How to protect the personal information of patients in the cloud environment has become an urgent problem to be solved. Medical image watermarking technology is an effective method to solve this problem. Combining the characteristics of Tent chaos and Henon chaos, this paper designed a Tent-Henon-Map double chaos watermarking encryption method and designed a medical image encryption watermarking algorithm based on ridgelet-DCT transform. The watermark images were encrypted by the Tent-Henon-Map double chaos which had the characteristics of sensitive initial values and large key space. Then, the feature vectors of the medical images were extracted through ridgelet-DCT transform. On the basis of ordinary watermarking technology, combined with zero watermarking, third-party concepts, and cryptographic technology, watermarking had a good ability to resist image processing attacks. The experimental results showed that the key space of the algorithm was {10}^{116}, which had better encryption and hard to crack. The time of watermark embedding and extraction were only 0.336 s and 0.439 s, with lower computational cost. And under high-strength conventional attacks and geometric attacks, the NC values of the algorithm were all greater than 0.55, which could effectively extract watermark information. It shown that the algorithm proposed had good robustness against conventional and geometric attacks It shown that the algorithm proposed had good robustness against conventional and geometric attacks, while taking into account the security.

Classification of Malware Families Based on Efficient-Net and 1D-CNN Fusion

A malware family classification method based on Efficient-Net and 1D-CNN fusion is proposed. Given the problem that some local information of malware itself as one-dimensional data will be lost when the malware is imaged, the malware is converted into an image and one-dimensional vector and then input into two neural networks. The network of two-dimensional convolution architecture is used to extract the texture features of malware, and the one-dimensional convolution is used to extract the features of local adjacent information, the deep characteristics of different networks are fused, and the two networks are modified at the same time during backpropagation. This method not only extracts the texture features of malware but also saves the features of the malware itself as one-dimensional data, which shows better performance for multiple datasets.

SAR Target Recognition via Monogenic Signal and Gaussian Process Model

The monogenic signal and Gaussian process model are applied to synthetic aperture radar (SAR) target recognition. The monogenic signal is used to extract the features of the SAR image. The Gaussian process model is a statistical learning algorithm based on the Bayesian theory, which constructs a classification model by combining the kernel function and the probability judgement. Compared with the traditional classification model, the Gaussian process model can obtain higher classification efficiency and accuracy. During the implementation, the monogenic feature vector of the SAR image is used as the input, and the target label is used as the output to train the Gaussian process model. For the test sample to be classified, the target label is determined by calculating the posterior probability of each class using the Gaussian process model. In the experiments, the validations are carried out under typical conditions based on the MSTAR dataset. According to the experimental results, the proposed method maintains the highest performance under the standard operating condition, depression angle differences, and noise corruption, which verifies its effectiveness and robustness.

Simulation of Video Image Fault Tolerant Coding Transmission in Digital Multimedia

In order to effectively improve the quality of video image transmission, this paper proposes a method of digital multimedia video image coding. The transmission of digital multimedia video image fault-tolerant coding requires sparse decomposition of a digital multimedia video image to obtain the linear form of the image and complete the transmission of video image fault-tolerant coding. The traditional method of fault-tolerant coding is based on human visual characteristics but ignores the linear form of the digital multimedia video image, which leads to the unsatisfactory effect of coding and transmission. In this paper, a fault-tolerant coding method based on wavelet transform and vector quantization is proposed to decompose and reconstruct digital multimedia video images. The smoothness of wavelet transform can remove visual redundancy; the decomposed image is vector quantized. The mean square deviation method and the similar scalar optimal quantization method are used to select and calculate the image vector, construct the over complete database of a digital multimedia video image, and normalize it; the digital multimedia video image is thinly decomposed by asymmetric atoms, and a linear representation of the image is obtained. According to the above-given operations, we can master the distribution range and law of pixels and realize fault-tolerant coding. The experimental results show that when the number of iterations is 15, the CR index is the same, PSNR increases by 8.7%, coding is 23.7% faster and decoding is 15% faster. Conclusion. The proposed method can not only improve the speed of fault-tolerant coding but also improve the quality of video image transmission.

GAN based simultaneous localization and mapping framework in dynamic environment

ObjectiveBecause of interference from dynamic objects, the traditional simultaneous localization and mapping (SLAM) framework execute poorly while operational in a dynamic environment. The Dynamic-SLAM model is introduced by considering the merits of deep learning in object discovery. At the semantic level, the dynamic objects in the new detection are detected to construct the prior knowledge with an SSD object detector. Among the radar and the LiDAR, the synchronization and conversion calibration makes the multi-sensor fusion. Localizing using a camera in a dynamic environment is more challenging because the localization process takes place in moving segments. This further results in unstable pose estimation. A large amount of information of the visual point cloud and high precision of the laser radar information enhances the accuracy of real-time positioning thereby attaining grid map and 3-D point cloud map. MethodsHere, we have proposed Generative Adversarial Network (GAN) with Aquila Optimizer for moving object detection. The LiDAR measurements check the Radar outcomes. The targeted moving objects are determined via Doppler velocity from the radar and their exact location and mass can be estimated with LiDAR and the proposed GAN-AO approach. Hence the GAN-based AO approach is used to segment the objects inside the point clouds. The arrangements of point clouds are made in a particular range to multiply the vertical points with the laser channel number. If the same objects are identified then the angles are analyzed among the image vectors and then labeled identically for the same points of the object. In addition to this, Velocity compensation is made to estimate the actual moving target from the world frame. This is due to the fact that the estimated velocity of the mmW-radar is in the radial direction with respect to the touching sensors. ResultsThe investigation is conducted between different state-of-art methods like Deep Learning (DL), Generative Adversarial Network (GAN), artificial neural network (ANN), Deep neural network (DNN) and propose methods. From this analysis, the proposed method provided 93.8% detection accuracy than other existing methods like DL, ANN, GAN and DNN respectively. ConclusionsWhile comparing to the state-of-art techniques, the proposed method demonstrated superior performance results in terms of tracking, detection, root mean square error (RMSE) and accuracy.

Machine Learning in Alzheimer’s Disease: Prognostic Prediction via Neuroimaging and Numerical Data

Alzheimer's disease (AD) is often detected too late or inaccurately in clinical practice. Therefore, improvement in the current methods of AD detection will provide opportunities for early intervention, symptomatic treatment, and, overall, better quality of patients’ and their caregivers’ lives. The paper is an in-depth study of how functional brain imaging and support vector machine (SVM) could be utilized to detect the risk for AD which works by assessing and plotting data into multi-dimensional graphs for various results. It aims to identify patients in presymptomatic stages for early treatment to delay or prevent progressive cognitive decline and disease. With knowledge of machine learning, our medical tool is a breakthrough in the methodology of AD detection. The future of our tool requires a substantive amount of brain scan data for the machine learning algorithm to produce reliable results, so further research in this field of study is crucial and strongly encouraged. 

Morpholine-Functionalized Multicomponent Metallacage as a Vector for Lysosome-Targeted Cell Imaging

Metallacages with suitable cavities and specific functions are promising delivery vectors in biological systems. Herein, we report a morpholine-functionalized metallacage for lysosome-targeted cell imaging. The efficient host-guest interactions between the metallacage and dyes prevent them from aggregation, so their emission in aqueous solutions is well maintained. The fluorescence quantum yield of these host-guest complexes reaches 74.40%. Therefore, the metallacage is further employed as a vector to deliver dyes with different emission colors (blue, green, and red) into lysosomes for targeted imaging. This research affords a type of vector for the delivery of various cargos toward biological applications, which will enrich the usage of metallacages in biomedical engineering.

Attention-based multiple-instance learning for Pediatric bone age assessment with efficient and interpretable

Pediatric bone age assessment (BAA) is a common clinical technique for evaluating children's endocrine, genetic, and growth disorders. However, the deep learning BAA method based on global images neglects fine-grained concerns, and regions of interest (ROIs) need additional annotation and complex processing. To overcome these shortcomings, we proposed an interpretable deep-learning architecture based on multiple-instance learning to address BAA efficiently without additional annotations. We cropped the entire image into small patches and got patch features by feature extraction network. Then, an attention backbone ranked feature vectors of the entire image and aggregates its information according to its relative importance. Finally, each image's features and gender were aggregated to predict bone age. The proposed method can identify ROIs by attention-based multi-instance aggregation without additional labels and produce interpretable heatmaps. Moreover, by cropping the complete image into patches and reducing the dimensionality, the proposed model can notice the fine-grained information of the image and improve the model training speed. We validated the proposed method in the Radiological Society of North America 2017 dataset. The results showed that the proposed model achieved an advanced performance of MAE 4.17 months. Furthermore, the visualization results indicated that the proposed model was highly interpretable, which can localize the ROIs without spatial labeling. In conclusion, a novel method for high performance and interpretable bone age prediction without additional manual annotations has been developed, which can be used to effectively assess the pediatric's bone age.

Development of Fibroblast Activation Protein Inhibitor-Based Dimeric Radiotracers with Improved Tumor Retention and Antitumor Efficacy.

Fibroblast activation protein (FAP), a fundamental component of the tumor stroma, is overexpressed in cancer-associated fibroblasts (CAFs). As a promising theranostic probe, we evaluated whether the FAP inhibitor (FAPI) dimer (DOTA-2P[FAPI]2) is more effective than its monomeric analogs for FAP-targeted radionuclide therapy. [68Ga]Ga/[177Lu]Lu-DOTA-2P(FAPI)2 were assayed in a stability study, small-animal positron emission tomography (PET) and single-photon emission computed tomography (SPECT), biodistribution, and radionuclide therapy to comprehensively evaluate their preclinical pharmacokinetics. The pharmacokinetics of [68Ga]Ga-DOTA-2P(FAPI)2 and [177Lu]Lu-DOTA-2P(FAPI)2 were determined in FAP-positive hepatocellular carcinoma patient-derived xenografts (PDXs) and HT-1080-FAP cell-derived xenografts (CDXs). [68Ga]Ga-DOTA-2P(FAPI)2 and [177Lu]Lu-DOTA-2P(FAPI)2 were stable in phosphate-buffered saline for 4 h. The tumor retention of [68Ga]Ga-DOTA-2P(FAPI)2 was better than that of [68Ga]Ga-FAPI-46 in HT-1080-FAP CDXs, while healthy organs showed low tracer uptake and fast body clearance. In single-photon emission computed tomography, [177Lu]Lu-DOTA-2P(FAPI)2 showed a higher uptake and longer retention for tumors in both PDXs and CDXs from 1-48 h. [177Lu]Lu-DOTA-2P(FAPI)2 showed the best inhibition of tumor growth in PDXs and CDXs. DOTA-2P(FAPI)2 has increased tumor uptake and retention properties compared to FAPI-46, which significantly improves the use of FAPI-based vectors for PET imaging and radionuclide therapy. [177Lu]Lu-DOTA-2P(FAPI)2 may be safe and effective for the treatment of FAP-positive malignant tumors.

Comparative analysis of classifiers for face recognition on image fragments identified by the FaceNet neural network

As a result of the analysis of the literature, the based methods of face recognition on fragments of color images were identified. These are flexible comparison in graphs, hidden Markov models, principal component analysis, and neural network methods. The analyzed methods of face recognition are mainly characterized by significant computational costs and low recognition performance. An exception is the neural network methods of face recognition, which, after completing the training, make it possible to obtain a high recognition performance at low computational costs. However, when changing the prototype images of faces, it often becomes necessary to redefine the network architecture and retrain the network. The specificity of neural network methods is also the complexity of selecting the network architecture and its training. Such papers are devoted to the use of neural networks onlyfor extraction of feature vectors of face images. The classification of the obtained feature vectors is then performed by knownmethods, namely, thresholding, a linear support vector machine, nearest neighbors, random forest. It has been observed that the lighting conditions in which the images were obtained and the turning of the head affect the shape of the separating surface and can decrease the feature vector classification performance for face images. Therefore, to improve the classification performance, it was decided to use correlation for prototype matching, a non-linear support vector machine and logistic regression. The performed experiment showed that correlation for prototype matching of low-light face images is characterized by higher classification performance compared to the thresholding. Moreover, the use of the Pearson and Spearman correlation coefficients showed similar results, and when using the Kendall correlation coefficient, the worst classification performance was obtained compared to the Pearson andSpearman coefficients. The research of the classification performance of images of faces that differ in head turn using correlation for prototype matching, a non-linear support vector machine and logistic regression showed the following. Correlation for prototype matching is more appropriate to use with small amounts of data due to the high classification performance and low computational complexity, since a small amount of data does not require a significant number of comparisons. However, on large amounts of data, the non-linear support vector machine requires less computation and shows similar classification performance. Using the results of the experiment, the researcher can select classification methods for a specific set of face images, preliminarily representing them with feature vectors using the network FaceNet.

Large-area total-thickness imaging and Burgers vector analysis of dislocations in β -Ga2O3 using bright-field x-ray topography based on anomalous transmission

Using bright-field x-ray topography based on anomalous transmission (AT), we have demonstrated the first large-area total-thickness imaging of dislocations in β-Ga2O3 at the substrate scale. The dislocation images were acquired from the entire 10 mm × 15 mm × 680 μm (001)-oriented substrate prepared by edge-defined film-fed growth (EFG) by stitching together hundreds of topographic images, each recorded with the forward-diffracted beam in the Laue geometry for g = 020, 0–20, 022, and 400, under the conditions in which AT occurred. Dislocations distributed over the entire crystal volume were imaged as long as their Burgers vectors (b) were not orthogonal to the g-vectors. The results of the g·b analysis of the dislocation contrasts clearly revealed three major dislocation types that were numerically dominant in the EFG crystal: (i) b-axis screw-type dislocations with b∥ξ∥[010] (ξ is the unit vector of line direction), (ii) b-axis edge-type dislocations with b∥[001] and ξ∥[010], and (iii) curved mixed-type dislocations lying on the (001) planes with b∥[010]. Based on their b- and ξ-vectors, types (i) and (ii) were attributed to dislocations that propagated during EFG pulling up along the [010] direction, while type (iii) was attributed to dislocations generated through glide in the [010](001) slip system under stress. The extent to which AT can manifest itself is explained by using the effective absorption coefficient calculated for the above g-vectors based on dynamical x-ray diffraction theory.

Performance Improvement of Face Recognition Method and its Application in COVID-19

In this paper an efficient face recognition technique is presented by integrating Discrete Wavelet Transform and Compressive Sensing based classifier. At first discrete wavelet transform has been applied on each face images. Then an image fusion technique has been applied on the decomposed image to provide better detail information of face images. Principal component Analysis is applied on fused face images to extract the feature vector. Finally, the feature vector of test images is extracted and classified by Compressive Sensing based Classifier. This proposed technique is also tested on two publicly available databases on AR and ORL. This technique is also tested on masked face images and experimental result shows improved performance compared to conventional PCA.

Identifying Depression in Parkinson's Disease by Using Combined Diffusion Tensor Imaging and Support Vector Machine.

ObjectiveTo investigate white matter microstructural alterations in Parkinson's disease (PD) patients with depression using the whole-brain diffusion tensor imaging (DTI) method and to explore the DTI–based machine learning model in identifying depressed PD (dPD).MethodsThe DTI data were collected from 37 patients with dPD and 35 patients with non-depressed PD (ndPD), and 25 healthy control (HC) subjects were collected as the reference. An atlas-based analysis method was used to compare fractional anisotropy (FA) and mean diffusivity (MD) among the three groups. A support vector machine (SVM) was trained to examine the probability of discriminating between dPD and ndPD.ResultsAs compared with ndPD, dPD group exhibited significantly decreased FA in the bilateral corticospinal tract, right cingulum (cingulate gyrus), left cingulum hippocampus, bilateral inferior longitudinal fasciculus, and bilateral superior longitudinal fasciculus, and increased MD in the right cingulum (cingulate gyrus) and left superior longitudinal fasciculus-temporal part. For discriminating between dPD and ndPD, the SVM model with DTI features exhibited an accuracy of 0.70 in the training set [area under the receiver operating characteristic curve (ROC) was 0.78] and an accuracy of 0.73 in the test set (area under the ROC was 0.71).ConclusionDepression in PD is associated with white matter microstructural alterations. The SVM machine learning model based on DTI parameters could be valuable for the individualized diagnosis of dPD.

Evaluation and selection of a lead diabody for interferon-γ PET imaging

IntroductionInterferon-γ (IFN-γ) is an appealing target to evaluate immune response in cancer immunotherapy as it is a hallmark of an active immune system. Imaging and detection via immunopositron emission tomography (immunoPET) of this soluble cytokine has been made feasible using a 89Zr-labeled (t 1/2 ~ 3.27 d) monoclonal antibody (mAb). Because of its size, using a full-length mAb as an imaging vector is not ideal for repeat serial imaging because of its prolonged blood pool residency and tumor accumulation resulting in lengthier wait times between administration and imaging. This consequently impacts the potential to image a dynamic immune response in real time. This work compares 89Zr-labeled diabodies (Db) designed with variable linker lengths between the VH and VL regions with the goal of selecting a lead Db for future studies. Methods and resultsFour Db fragments with various linker lengths (HL-n, n = 7–13 amino acids) were each conjugated to desferrioxamine (DFO). The number of attached chelates was analyzed via mass spectrometry with all immunoconjugates exhibiting one unit of DFO attached. Db-DFO conjugates were subsequently radiolabeled with zirconium-89. All constructs radiolabeled with high yields. Each radioimmunoconjugate was tested for reactivity to IFN-γ. All tracers except for [89Zr]Zr-DFO-NCS-anti-IFN-γ HL-9 exhibited comparable immunoreactivities (>90 %) to the radiolabeled parent mAb (95.8 %). At 24 h post-labeling, the IRF values were retained except for the HL-13 construct. Imaging scans and tissue distribution studies acquired in mice bearing CT26 syngeneic colorectal tumors between 1 and 24 h post-tracer administration demonstrated variable clearance kinetics and tumor localization of each radiotracer. HL-7 had higher binding in non-tumor tissues compared to HL-11 and HL-13 at 3 h p.i. Competitive binding studies versus unmodified parent mAb (AN-18) demonstrated blocking of radiolabeled HL-11 and HL-13. [89Zr]Zr-DFO-NCS-anti-IFN-γ HL-7 was inadequately blocked. ConclusionDespite nuanced differences in linker lengths, our data demonstrates that [89Zr]Zr-DFO-NCS-anti-IFN-γ HL-11 exhibited the best radiotracer properties for the assessment of IFN-γ production in vivo. Work is currently underway to test the potential of using shorter-lived isotopes, like copper-64 (t1/2 ~ 12.7 h) to match pharmacokinetics and half-lives.

TCB-spline-based Image Vectorization

Vector image representation methods that can faithfully reconstruct objects and color variations in a raster image are desired in many practical applications. This article presents triangular configuration B-spline (referred to as TCB-spline)-based vector graphics for raster image vectorization. Based on this new representation, an automatic raster image vectorization paradigm is proposed. The proposed framework first detects sharp curvilinear features in the image and constructs knot meshes based on the detected feature lines. It iteratively optimizes color and position of control points and updates the knot meshes. By using collinear knots at feature lines, both smooth and discontinuous color variations can be efficiently modeled by the same set of quadratic TCB-splines. A variational knot mesh generation method is designed to adaptively introduce knots and update their connectivity to satisfy the local reconstruction quality. Experiments and comparisons show that our framework outperforms the existing state-of-the-art methods in providing more faithful reconstruction results. In particular, our method is able to model undetected features and subtle or complicated color variations in-between features, which the previous methods cannot handle efficiently. Our vectorization representation also facilitates a variety of editing operations performed directly over vector images.

Foreword to the Special Section on Smart Tools and Applications in Graphics (STAG 2021)

The Special Section contains extended and revised versions of the best papers presented at the 8th Conference on Smart Tools and Applications in Graphics (STAG 2021), held virtually on October 28–29, 2021. Five papers were selected by appointed members from the Program Committee; extended versions were submitted and further reviewed by external experts. The result is a collection of paper spanning different Visual Computing domains: from vector graphics on surfaces to image vectorization, from 3D geometric modelling of geological data to computation of the geometric kernel of polyhedra, up to lens-based scene exploration of annotated visual data.

The effect of spatial sampling on magnetic field modeling and helicity computation

Context. Nonlinear force-free (NLFF) modeling is regularly used to indirectly infer the 3D geometry of the coronal magnetic field, which is not otherwise accessible on a regular basis by means of direct measurements. Aims. We study the effect of binning in time-series NLFF modeling of individual active regions (ARs) in order to quantify the effect of a different underlying spatial sampling on the quality of modeling as well as on the derived physical parameters. Methods. We apply an optimization method to sequences of Solar Dynamics Observatory (SDO) Helioseismic and Magnetic Imager (HMI) vector magnetogram data at three different plate scales for three solar active regions to obtain nine NLFF model time series. From the NLFF models, we deduce active-region magnetic fluxes, electric currents, magnetic energies, and relative helicities, and analyze those with respect to the underlying spatial sampling. We calculate various metrics to quantify the quality of the derived NLFF models and apply a Helmholtz decomposition to characterize solenoidal errors. Results. At a given spatial sampling, the quality of NLFF modeling is different for different ARs, and the quality varies along the individual model time series. For a given AR, modeling at a certain spatial sampling is not necessarily of superior quality compared to that performed with a different plate scale. Generally, the NLFF model quality tends to be higher for larger pixel sizes with the solenoidal quality being the ultimate cause for systematic variations in model-deduced physical quantities. Conclusions. Optimization-based modeling using SDO/HMI vector data binned to larger pixel sizes yields variations in magnetic energy and helicity estimates of ≲30% on overall, given that concise checks ensure the physical plausibility and high solenoidal quality of the tested model. Spatial-sampling-induced differences are relatively small compared to those arising from other sources of uncertainty, including the effects of applying different data calibration methods, those of using vector data from different instruments, or those arising from application of different NLFF methods to identical input data.

Similar goods search based on FAISS

Problem. The problem of obtaining a complete set of similar goods from different manufacturers based on the image and description of the product is relevant and interesting. The article includes description of algo and architecture solution for online goods clustering. Goal. The goal of the work is to create and explore models of e-commerce ML (online goods clustering). The work includes developing AI components based on FAISS and Deep Learning to implement Product Quantization of goods searches based on their embedding vectors and vectors of images. The proposed approach to the search for similar goods uses the DNN for features detection and embedding vectors, which are used for FAISS clustering. Methodology. The analytical and empirical methods of research based on the development and DNN are used, ML methods to determine features of goods and solve classification problems. Results. The architecture of the solution is based on the use of GCP services. Practical value. AI solutions have practical value for e-commerce, they help retain or return customers, and recommend products. As a result of the work, a basic solution to the problem of matching similar products based on the use of FAISS and Deep Learning algorithms in GCP was developed and tested.

Research on High-Resolution Face Image Inpainting Method Based on StyleGAN

In face image recognition and other related applications, incomplete facial imagery due to obscuring factors during acquisition represents an issue that requires solving. Aimed at tackling this issue, the research surrounding face image completion has become an important topic in the field of image processing. Face image completion methods require the capability of capturing the semantics of facial expression. A deep learning network has been widely shown to bear this ability. However, for high-resolution face image completion, the network training of high-resolution image inpainting is difficult to converge, thus rendering high-resolution face image completion a difficult problem. Based on the study of the deep learning model of high-resolution face image generation, this paper proposes a high-resolution face inpainting method. First, our method extracts the latent vector of the face image to be repaired through ResNet, then inputs the latent vector to the pre-trained StyleGAN model to generate the face image. Next, it calculates the loss between the known part of the face image to be repaired and the corresponding part of the generated face imagery. Afterward, the latent vector is cut to generate a new face image iteratively until the number of iterations is reached. Finally, the Poisson fusion method is employed to process the last generated face image and the face image to be repaired in order to eliminate the difference in boundary color information of the repaired image. Through the comparison and analysis between two classical face completion methods in recent years on the CelebA-HQ data set, we discovered our method can achieve better completion results of 256*256 resolution face image completion. For 1024*1024 resolution face image restoration, we have also conducted a large number of experiments, which prove the effectiveness of our method. Our method can obtain a variety of repair results by editing the latent vector. In addition, our method can be successfully applied to face image editing, face image watermark clearing and other applications without the network training process of different masks in these applications.

Vigenere and genetic cross-over acting at the restricted ASCII code level for color image encryption.

In the current high-tech era, the growing popularity of e-healthcare has heightened the attention focused on data security against cyber-attacks. As digital medical images are transported over the public network, an intermediate degree of protection must be in place. One major technique is the implementation of encryption to safeguard medical images. This article recommends a new medical image encryption technology. This algorithm began with a simple application of advanced Vigenere technology, followed by the implementation of a deeply adapted genetic operator for color image encryption, with the intention of better integrating into the medical image encryption security sector. After vectorization of the original image, the first round of Vigenere control by two chaotic substitution tables attached to highly modified replacement functions is applied. The output vector is encoded in restricted ASCII code in order to obtain a genetic hybridization suitable for encryption of medical and color images. A strong chaining will be applied between the encrypted block and the following clear block in order to increase the impact of the avalanche effect and to protect the system against any differential attack. Simulations carried out on a large volume of images of different sizes and formats ensure that our approach is not exposed to known attacks. HIGHLIGHTS: The two rounds of our algorithm represent a large modification of a classical and a genetic operator. The main new features are ✓ First round ○ Vigenere upgrade ○ Selected simulations from the primary round ✓ Switching to reduced ASCIIcode ○ Two cross-over tables building process ○ Chaotic operation ASCII table ○ Output vector encryption ■ Detailed simulation study ■ Benchmarking with other recent algorithms.