False Positive Problem Research Articles

A novel Deep High-level Concept-mining Jointing Hashing Model for unsupervised cross-modal retrieval

Unsupervised cross-modal hashing has achieved great success in various information retrieval applications owing to its efficient storage usage and fast retrieval speed. Recent studies have primarily focused on training the hash-encoded networks by calculating a sample-based similarity matrix to improve the retrieval performance. However, there are two issues remain to solve: (1) The current sample-based similarity matrix only considers the similarity between image-text pairs, ignoring the different information densities of each modality, which may introduce additional noise and fail to mine key information for retrieval; (2) Most existing unsupervised cross-modal hashing methods only consider alignment between different modalities, while ignoring consistency between each modality, resulting in semantic conflicts. To tackle these challenges, a novel Deep High-level Concept-mining Jointing Hashing (DHCJH) model for unsupervised cross-modal retrieval is proposed in this study. DHCJH is able to capture the essential high-level semantic information from image modalities and integrate into the text modalities to improve the accuracy of guidance information. Additionally, a new hashing loss with a regularization term is introduced to avoid the cross-modal semantic collision and false positive pairs problems. To validate the proposed method, extensive comparison experiments on benchmark datasets are conducted. Experimental findings reveal that DHCJH achieves superior performance in both accuracy and efficiency. The code of DHCJH is available at Github.

Establishment of LAMP-CRISPR/Cas12a for rapid detection of Escherichia coli O157:H7 and one-pot detection

Escherichia coli O157:H7 is a pathogenic serotype of Escherichia coli. Consumption of food contaminated with E. coli O157:H7 could cause a range of diseases. Therefore, it is of great importance to establish rapid and accurate detection methods for E. coli O157:H7 in food. In this study, based on LAMP and combined with the CRISPR/cas12a system, a sensitive and specific rapid detection method for E. coli O157:H7 was established, and One-Pot detection method was also constructed. The sensitivity of this method could stably reach 9.2 × 10° CFU/mL in pure culture, and the whole reaction can be completed within 1 h. In milk, E. coli O157:H7 with an initial contamination of 7.4 × 10° CFU/mL only needed to be cultured for 3 h to be detected. The test results can be judged by the fluorescence curve or by visual observation under a UV lamp, eliminating instrument limitations and One-Pot detection can effectively prevent the problem of false positives. In a word, the LAMP-CRISPR/cas12a system is a highly sensitive and convenient method for detecting E. coli O157:H7.

Reverse network diffusion to remove indirect noise for better inference of gene regulatory networks.

Gene regulatory networks (GRNs) are vital tools for delineating regulatory relationships between transcription factors and their target genes. The boom in computational biology and various biotechnologies has made inferring GRNs from multi-omics data a hot topic. However, when networks are constructed from gene expression data, they often suffer from false-positive problem due to the transitive effects of correlation. The presence of spurious noise edges obscures the real gene interactions, which makes downstream analyses, such as detecting gene function modules and predicting disease-related genes, difficult and inefficient. Therefore, there is an urgent and compelling need to develop network denoising methods to improve the accuracy of GRN inference. In this study, we proposed a novel network denoising method named REverse Network Diffusion On Random walks (RENDOR). RENDOR is designed to enhance the accuracy of GRNs afflicted by indirect effects. RENDOR takes noisy networks as input, models higher-order indirect interactions between genes by transitive closure, eliminates false-positive effects using the inverse network diffusion method, and produces refined networks as output. We conducted a comparative assessment of GRN inference accuracy before and after denoising on simulated networks and real GRNs. Our results emphasized that the network derived from RENDOR more accurately and effectively captures gene interactions. This study demonstrates the significance of removing network indirect noise and highlights the effectiveness of the proposed method in enhancing the signal-to-noise ratio of noisy networks. The R package RENDOR is provided at https://github.com/Wu-Lab/RENDOR and other source code and data are available at https://github.com/Wu-Lab/RENDOR-reproduce.

Modeling of fabric sewing break detection based on U-Net network

To solve the problem of false positives and false negatives in the manual detection of fabric sewing breaks, a method of fabric sewing break detection based on the U-Net network is proposed. By detecting the adjacent distance between the characteristic contours of adjacent sewing stitches, the distribution uniformity of sewing stitches in sewing patterns is calculated, and the abnormal detection and traceability of fabric sewing broken threads are realized. First, the U-Net network sewing feature extraction model was trained using sewing images and their corresponding stitching feature annotation maps. Then, the trained network model was used to process sewing image samples to obtain binary stitching feature maps. Second, the stitching feature maps were processed using a closing operation to eliminate residual image noise. On this basis, the template matching algorithm was used to extract the stitching feature contours. Finally, according to the distance between adjacent feature contours, the fabric sewing break detection and abnormality tracing model was constructed. The model is validated by examples, and the results show that the abnormal samples of stitching lines are detected, and the corresponding break positions are given. The overall detection accuracy of the model is 95.75%, indicating that the constructed fabric sewing break detection model is effective.

A Novel Deep Learning Method for Detecting Strawberry Fruit

The recognition and localization of strawberries are crucial for automated harvesting and yield prediction. This article proposes a novel RTF-YOLO (RepVgg-Triplet-FocalLoss-YOLO) network model for real-time strawberry detection. First, an efficient convolution module based on structural reparameterization is proposed. This module was integrated into the backbone and neck networks to improve the detection speed. Then, the triplet attention mechanism was embedded into the last two detection heads to enhance the network’s feature extraction for strawberries and improve the detection accuracy. Lastly, the focal loss function was utilized to enhance the model’s recognition capability for challenging strawberry targets, which thereby improves the model’s recall rate. The experimental results demonstrated that the RTF-YOLO model achieved a detection speed of 145 FPS (frames per second), a precision of 91.92%, a recall rate of 81.43%, and an mAP (mean average precision) of 90.24% on the test dataset. Relative to the baseline of YOLOv5s, it showed improvements of 19%, 2.3%, 4.2%, and 3.6%, respectively. The RTF-YOLO model performed better than other mainstream models and addressed the problems of false positives and false negatives in strawberry detection caused by variations in illumination and occlusion. Furthermore, it significantly enhanced the speed of detection. The proposed model can offer technical assistance for strawberry yield estimation and automated harvesting.

Adaptive Hardness Negative Sampling for Collaborative Filtering

Negative sampling is essential for implicit collaborative filtering to provide proper negative training signals so as to achieve desirable performance. We experimentally unveil a common limitation of all existing negative sampling methods that they can only select negative samples of a fixed hardness level, leading to the false positive problem (FPP) and false negative problem (FNP). We then propose a new paradigm called adaptive hardness negative sampling (AHNS) and discuss its three key criteria. By adaptively selecting negative samples with appropriate hardnesses during the training process, AHNS can well mitigate the impacts of FPP and FNP. Next, we present a concrete instantiation of AHNS called AHNS_{p

Siamese Transformer-Based Building Change Detection in Remote Sensing Images.

To address the challenges of handling imprecise building boundary information and reducing false-positive outcomes during the process of detecting building changes in remote sensing images, this paper proposes a Siamese transformer architecture based on a difference module. This method introduces a layered transformer to provide global context modeling capability and multiscale features to better process building boundary information, and a difference module is used to better obtain the difference features of a building before and after a change. The difference features before and after the change are then fused, and the fused difference features are used to generate a change map, which reduces the false-positive problem to a certain extent. Experiments were conducted on two publicly available building change detection datasets, LEVIR-CD and WHU-CD. The F1 scores for LEVIR-CD and WHU-CD reached 89.58% and 84.51%, respectively. The experimental results demonstrate that when utilized for building change detection in remote sensing images, the proposed method exhibits improved robustness and detection performance. Additionally, this method serves as a valuable technical reference for the identification of building damage in remote sensing images.

Retinal Layer Segmentation in OCT Images With Boundary Regression and Feature Polarization.

The geometry of retinal layers is an important imaging feature for the diagnosis of some ophthalmic diseases. In recent years, retinal layer segmentation methods for optical coherence tomography (OCT) images have emerged one after another, and huge progress has been achieved. However, challenges due to interference factors such as noise, blurring, fundus effusion, and tissue artifacts remain in existing methods, primarily manifesting as intra-layer false positives and inter-layer boundary deviation. To solve these problems, we propose a method called Tightly combined Cross-Convolution and Transformer with Boundary regression and feature Polarization (TCCT-BP). This method uses a hybrid architecture of CNN and lightweight Transformer to improve the perception of retinal layers. In addition, a feature grouping and sampling method and the corresponding polarization loss function are designed to maximize the differentiation of the feature vectors of different retinal layers, and a boundary regression loss function is devised to constrain the retinal boundary distribution for a better fit to the ground truth. Extensive experiments on four benchmark datasets demonstrate that the proposed method achieves state-of-the-art performance in dealing with problems of false positives and boundary distortion. The proposed method ranked first in the OCT Layer Segmentation task of GOALS challenge held by MICCAI 2022. The source code is available at https://www.github.com/tyb311/TCCT.

Loop closure detection of visual SLAM based on variational autoencoder.

Loop closure detection is an important module for simultaneous localization and mapping (SLAM). Correct detection of loops can reduce the cumulative drift in positioning. Because traditional detection methods rely on handicraft features, false positive detections can occur when the environment changes, resulting in incorrect estimates and an inability to obtain accurate maps. In this research paper, a loop closure detection method based on a variational autoencoder (VAE) is proposed. It is intended to be used as a feature extractor to extract image features through neural networks to replace the handicraft features used in traditional methods. This method extracts a low-dimensional vector as the representation of the image. At the same time, the attention mechanism is added to the network and constraints are added to improve the loss function for better image representation. In the back-end feature matching process, geometric checking is used to filter out the wrong matching for the false positive problem. Finally, through numerical experiments, the proposed method is demonstrated to have a better precision-recall curve than the traditional method of the bag-of-words model and other deep learning methods and is highly robust to environmental changes. In addition, experiments on datasets from three different scenarios also demonstrate that the method can be applied in real-world scenarios and that it has a good performance.

Color Image Watermarking Using Adaptive Tunable Q Wavelet, Finite Ridge Let, and Arnold Transforms with TSVD

In recent years, researchers have been working to integrate watermarking practices anchored in wavelet transform (WT) coefficients and Singular value decomposition (SVD) to recover the robustness of watermarked images in opposition to geometric and non-geometric attacks. Conversely, the watermark embedding, can initiate vagueness in the false-positive problem (FPP). To address this issue, Flamingo search algorithm-based Color Image Watermarking using Adaptive Tunable Q wavelet, Finite ridge let, Arnold transforms with Truncated Singular value decomposition (ATQWT-FRT-AT-TSVD-FSOA-CIW) is proposed. The proposed ATQWT-FRTAT-TSVD-FSOA-CIW method is separated into two stages: (i) watermark embedding and (ii) watermark extraction. The scaling factor is optimized with the help of the Flamingo search algorithm, which provides higher robustness, higher embedding strength, resistance to a false-positive problem, and better quality of the watermarking scheme. The proposed ATQWT-FRT-AT-TSVD-FSOA-CIW method is simulated in MATLAB tool. The performance of the proposed ATQWT-FRT-AT-TSVD-FSOA-CIW design is estimated under several performance metrics, such as Peak signal to noise ratio (PSNR), Structural similarity index measurement (SSIM), and Normalized cross-correlation (NCC). The performance of the proposed ATQWT-FRT-AT-TSVD-FSOA-CIW method attains 35%, 22.43%, 21.157%, 13.96%, and 7.186% higher PSNR compared to the existing designs, like secure with robust color image watermarking utilizing nature-inspired intelligence (DWT-SVD-ABCOA-CIW), the Artificial bee colony basis perceptually tuned blind color image watermarking in the hybrid LWT-DCT domain (LWT-DCT-ABCOA-CIW), the DWT-SVD based adaptive color multi-watermarking scheme for copyright protection under AMEF and PSO-GWO (DWT-SVD-Hybrid PSO-GWOA-CIW), Robust image watermarking based on fractional Krawtchouk transform with optimization (FrKT-Hybrid FO-CSOA-CIW), and the optimized semi-blind scheme for color image watermarking (DWTFRT-AT-SVD-PSO-CIW).

Feasibility of Gene Regulatory Network Inference Under Different Noise Conditions

One of the most difficult and pressing problems in computational cell biology is the inference of gene regulatory network structure from transcriptomic data. Although this topic has been widely studied for more than a decade, it is still unclear how well these methods work and how their performance differs under different conditions. We analyze the feasibility of network inference from mRNA abundance data using simulations of gene regulatory interactions, considering both intrinsic and extrinsic noise in the process of gene expression. We find that under conditions of only intrinsic noise in gene expression, the correlation between mRNA levels of genes in an activation relationship is quite low, suggesting that the task of network inference from transcriptomic data is very difficult under these conditions. By contrast, extrinsic noise affecting the process of gene expression, which could come from an upstream regulator, external stimulus, or change in the cell state, results in higher correlation between mRNA levels of these genes, potentially making the task of network inference from mRNA data more feasible. Lastly, we analyze the problem of false positives between genes that have no direct interaction but share a common upstream regulator, and explore a strategy for distinguishing between these false positives and true interactions based on noise profiles of mRNA expression levels.

Leveraging AI and Machine Learning for Cyber Threat Analysis

This paper explores how AI and Machine Learning is becoming a key participant in cybersecurity's emergence and constantly shifting nature. It disillusioned traditional forms of detection, such as signature-based detection and rule-based systems, with the idea that they are suitable for dealing with new complex threats like zero-day vulnerabilities, APTs, and social engineering attacks. AI and ML provide enhanced features for accurate time operation, real-time threat identification tools, anomaly identification tools, behavioral tools, and predictive analysis, and therefore, have become helpful in the modern context and central to the cybersecurity architecture of many organizations. The paper fleshes out the strengths of AI in intrusion detection, dynamic malware analysis, and phishing detection, among others. The following AI methods are covered in detail: Supervised and unsupervised learning algorithms are used in processing datasets, and autoencoders are used. Deep learning models are suitable for processing big data in real-time. The issues of AI incorporating false positive outcomes, adversarial AI, and data quality problems are also addressed, along with ethical and privacy considerations. The roles of automated incident reporting and self-learning security systems, as forecasted for AI's contributions in managing cyber threats, are designated as crucial. Finally, the paper concludes by stressing the need for organizations to adopt AI solutions for the preemptive approach in the cybersecurity domain.

Blind watermarking scheme for medical and non-medical images copyright protection using the QZ algorithm

The widely used watermarking technique is to hide the watermark information through adding process, which means the watermark is added with the host image to realize embedding. The constraints of these addition-based watermarking systems mainly are non-blind, they are vulnerable to forgery attacks and have the false-positive problem. To solve these problems, a novel blind watermarking scheme based on the QZ algorithm for medical and normal images is first presented in this paper. In the presented scheme, we take the watermark and the host image as two inputs of the QZ algorithm to realize embedding, rather than the adding process. The outputs of the QZ algorithm include four parts: two upper quasitriangular matrices and two unitary matrices. Then the three parts of the outputs are synthesized by the vector synthesis and inverse discrete wavelet transform-singular value decomposition (DWT-SVD) to obtain the watermarked host image. Due to the characteristics of the QZ algorithm, our scheme is blind. The performance is tested and compared by applying our scheme to both medical and non-medical images. The experimental results show that our scheme is robust against various image-processing attacks, forgery attacks, and false-positive problem. And from the comparison results, the proposed scheme outperforms the existing addition-based schemes in terms of robustness against the Median filter, Low-pass filter, Gaussian blur attacks, and forgery attacks, while maintaining good invisibility.

VSTNet: Robust watermarking scheme based on voxel space transformation for diffusion tensor imaging images

Deep watermarking has been successfully applied to real-value domain (e.g., CT, MRI, DWI) medical images for copyright protection; however, manifold-value domain images, such as diffusion tensor imaging (DTI), have received less academic attention. The non-Euclidean nature of DTI images hinders deep models from generating plausible DTI images because it is difficult to guarantee that each generated DTI image voxel lies on a 3 × 3 symmetric positive definite manifold. In this paper, we propose a robust watermarking scheme based on voxel space transformation for DTI images. First, a voxel space transformation is proposed, which effectively solves the problem of fiber orientation estimation in Riemann networks for DTI images and obtains the eigenvalues and eigenvectors of each voxel using singular value decomposition (SVD) to transform the DTI voxels from the 3 × 3 symmetric positive definite manifold space to Euclidean space. Second, SVD is combined with a deep neural network to extract the high-level features of each voxel eigenvalue in the DTI images and embedding watermarking messages, avoiding the false positive errors problem in traditional watermarking algorithms based on an SVD transform. Finally, using a deep neural network combining multiscale dilated convolution, dense residual connection and channel attention, the algorithm demonstrates high robustness against various intentional or unintentional attacks on DTI images while ensuring the good visual quality and diffusion characteristics of the DTI images embedded with watermarking messages. Experiments show that the watermarking message extraction accuracy reaches 97.8% for Crop (p = 0.7) attacks and 95.4% for Gaussian Blur (k = 7) attacks.

Deep matched filtering for retinal vessel segmentation

The structure of the retinal vascular tree can reflect many indicators of ophthalmic health status. Retinal vessel segmentation, an important basis for the quantitative analysis of vascular structure, has been developed over decades. However, in the areas of imaging quality and contrast, there are still vascular ruptures and missed detection by existing methods. Moreover, the problem of false positives caused by existing lesions has not been properly addressed. Inspired by the primary visual cortex of the biological visual system, this paper proposes Anisotropic Perceptive Convolution (APC) and Anisotropic Enhancement Module (AEM) to model visual cortex cells and their orientation selection mechanism, respectively, as well as a novel network named W-shaped Deep Matched Filtering (WS-DMF) having a W-shaped overall framework. In the feature extraction backbone, a DMF based on a multilayer aggregation of APC is designed to enhance vascular features and inhibit the expression of pathological features. In addition, AEMs are embedded into DMF to enhance the orientation and position information of high-dimensional features. Furthermore, in order to enhance the ability of APC for perceiving linear textures (such as blood vessels), an Orientation Anisotropic Loss (OAL) is introduced. Based on experiments on several widely used datasets and several professional vessel segmentation metrics, the proposed method exhibits strong abilities of vessel extraction and lesion inhibition, surpassing many state-of-the-art algorithms. The results indicate the potential value of WS-DMF in clinical practice. The source code is available at https://www.github.com/tyb311/WS-DMF.

The problem of false positives in automated census linking: Nineteenth-century New York’s Irish immigrants as a case study

Automated census linkage algorithms have become popular for generating longitudinal data on social mobility, especially for immigrants and their children. But what if these algorithms are particularly bad at tracking immigrants? This study utilizes a database on nineteenth-century Irish immigrants, generated from the most widely used algorithms, created by Abramitzky, Boustan, and Eriksson (ABE). Our objective is to assess the extent to which different individuals are erroneously linked together across census years and the consequences of these “false positives” for calculating social mobility. Our findings raise serious questions about the quality of the matches generated by the “first generation” of automated census linkage algorithms. False positives range from about one-third to one-half of all links. These bad links lead to sizeable estimation errors when measuring Irish immigrant social and geographic mobility.

Precise genome engineering in Pseudomonas using phage-encoded homologous recombination and the Cascade-Cas3 system.

A lack of generic and effective genetic manipulation methods for Pseudomonas has restricted fundamental research and utilization of this genus for biotechnology applications. Phage-encoded homologous recombination (PEHR) is an efficient tool for bacterial genome engineering. This PEHR system is based on a lambda Red-like operon (BAS) from Pseudomonas aeruginosa phage Ab31 and a Rac bacteriophage RecET-like operon (Rec-TEPsy) from P. syringae pv. syringae B728a and also contains exogenous elements, including the RecBCD inhibitor (Redγ or Pluγ) or single-stranded DNA-binding protein (SSB), that were added to enhance the PEHR recombineering efficiency. To solve the problem of false positives in Pseudomonas editing with the PEHR system, the processive enzyme Cas3 with a minimal Type I-C Cascade-based system was combined with PEHR. This protocol describes the utilization of a Pseudomonas-specific PEHR-Cas3 system that was designed to universally and proficiently modify the genomes of Pseudomonas species. The pipeline uses standardized cassettes combined with the concerted use of SacB counterselection and Cre site-specific recombinase for markerless or seamless genome modification, in association with vectors that possess the selectively replicating template R6K to minimize recombineering background. Compared with the traditional allelic exchange editing method, the PEHR-Cas3 system does not need to construct suicide plasmids carrying long homologous arms, thus simplifying the experimental procedure and shortening the traceless editing period. Compared with general editing systems based on phage recombinases, the PEHR-Cas3 system can effectively improve the screening efficiency of mutants using the cutting ability of Cas3 protein. The entire procedure requires ~12 days.

Dual-signal aptasensor based on Zr-MOF for ultrasensitive detection of AFB1 in corn

A dual-signal aptasensor for ultra-sensitive detection of AFB1 was developed based on the inherent electrochemical and fluorescence properties of Zr-MOF, leading to improved detection accuracy and reduced false-positive problems caused by individual signals. Using a large number of Zr (IV) metal sites on the surface of Zr-MOF, the signal recognition probe was constructed based on Lewis acid-base theory by forming a stable Zr-O-P ligand bond with the modified -PO43- at the end of the aptamer, and improving the binding stability and immobilization efficiency. The signal recognition probe Zr-MOF/Apt was constructed using this method and bound to cDNA which was anchored to the surface of highly porous gold electrode via the principle of base complementary pairing. When AFB1 presented, due to the aptamer binding it specifically, the signal probe fell off from the electrode surface into the electrolyte, resulting in a decrease in the electrochemical signal on the electrode, and at the same time, the fluorescence signal was increased in the electrolyte. The performance of the dual-signal aptasensor was confirmed by electrochemical and fluorescence signals and exhibited excellent detection performance under optimal conditions (linear range of 0.001–30 ng mL−1 and detection limit of 0.6 pg mL−1 for the electrochemical signal, 0.8 pg mL−1 for the fluorescence signal). It also showed good stability, selectivity, and reproducibility. The spiked recoveries in corn samples ranged from 95.0 % to 120.0 %.

A secure fractional-order polar harmonic transforms based image watermarking scheme using funk singular value decomposition

In addition to existing cryptographic systems, watermarking technologies have been developed to add extra security. Digital watermarking utilizes embedding or hiding techniques to protect multimedia files from copyright violations. Fundamental procedures of digital watermarking techniques are embedding and extraction. Singular value decomposition (SVD) based Image watermarking schemes become popular owing to its better trade-off among robustness and imperceptibility. Nevertheless, false positive problem (FPP) is a major issue of SVD-based watermarking schemes. The singular value that is a fixed value and does not contain structural information about image is the primary cause of FPP problem. Therefore, Message Digest algorithm image watermarking scheme based on Funk Singular Value Decomposition and Fractional-Order Polar Harmonic Transform (FSVD-FOPHT) is proposed in this paper to address this problem. The MD-5 algorithm is used to extract data from the host and watermark imageries and then create secret key. The FSVD-FOPHT method is utilized to hide watermark information in host image. The secret keys are extracted from hided image using inverse process of Fractional-Order Polar Harmonic Transforms with Funk Singular Value Decomposition algorithm. By using the extraction procedure, watermark image is extracted, and then reconstructs original watermarked image. During extraction procedure, the secret key is used for authentication to address FPP. Then, the proposed method is implemented in MATLAB and performance is analyzed with evaluation metrics, such as Embedding capacity, MSE, PSNR, and NC. The proposed method provide 14.6%, 17.34%, 19.53%, 21.46% and 23.89% high PSNR for cold-snow-landscape-water test image, 14.29%, 16.47%, 18.39%, 20.16% and 21.93% high PSNR for landscape-nature-sky-blue Test image, 16.85%, 19.99%, 22.70%, 27.22% and 29.16% high Embedding Capacity for cold-snow-landscape-water test image 22.83%, 24.64%, 27.92%, 29.60% and 31.77% high Embedding Capacity for landscape-nature-sky-blue Test image 35.38%, 32.63%, 30.95%, 28.61% and 26.08% low extraction time compared with existing methods SVD-CMSF-SIW, FE-IWS-DNN, AR-IWS-DNN, BBET-SHA1-SIW and LSB-DWT-SIW respectively.

Chain hybridization‐based CRISPR‐lateral flow assay enables accurate gene visual detection

Visualized gene detection based on the CRISPR‐Cas12/CRISPR‐Cas13 technology and lateral flow assay device (CRISPR‐LFA) has shown great potential in point‐of‐care testing sector. Current CRISPR‐LFA methodology mainly utilizes conventional immuno-based LFA test strips, which could visualize whether the reporter probe is trans-cleaved by Cas protein, indicating the target positive detection. However, conventional CRISPR‐LFA usually produces false-positive results in target negative assay. Herein, a nucleic acid Chain Hybridization‐based Lateral Flow Assay platform, named CHLFA, has been developed to achieve the CRISPR‐CHLFA concept. Different from the conventional CRISPR‐LFA, the proposed CRISPR‐CHLFA system was established based on the nucleic acid hybridization between the GNP‐probe embedded in test strips and ssDNA (or ssRNA) reporter from CRISPR (LbaCas12a or LbuCas13a) reaction, which eliminated the requirement of immunoreaction in conventional immuno-based LFA. The assay realized the detection of 1–10 copy of target gene per reaction within 50 min. The CRISPR-CHLFA system achieved highly accurate visual detection of target negative samples, thus overcoming the false-positive problem that often produced in assays using conventional CRISPR‐LFA. The CRISPR‐CHLFA platform was further adopted for the visual detection of marker gene from SASR‐CoV-2 Omicron variant and Mycobacterium tuberculosis (MTB), respectively, and 100% accuracy for the analysis of clinical specimens (45 SASR‐CoV-2 specimens and 20 MTB specimens) was obtained. The proposed CRISPR-CHLFA system could provide an alternative platform for the development of POCT biosensors and can be widely adopted in accurate and visualized gene detection.