10 results about "Image fusion algorithm" patented technology

This invention discloses a visual inspection method and apparatus for glass production quality, relating to the field of glass quality inspection technology. The method includes the following steps: constructing a polarization excitation detection field; acquiring and registering a sequence of original polarization images of the glass based on the polarization excitation detection field to obtain a polarization image sequence; performing image fusion on the polarization image sequence using an image fusion algorithm based on non-subsampled contour wave transform, sparse representation, and guided filters to obtain a defect-enhanced image; inputting the defect-enhanced image into a lightweight glass defect recognition model based on improved self-attention, outputting a glass defect category, acquiring glass defect quality impact data, and classifying glass quality grades based on the glass defect category and the glass defect quality impact data, thereby achieving visual inspection of glass production quality.

This invention relates to the field of inkjet characteristic measurement technology, specifically to inkjet characteristic measurement methods, systems, equipment, and storage media. A host industrial control computer ensures efficient inkjet start-up control, while a printhead driver board ensures stable, accurate, and timely inkjet printing. A camera device captures images via commands for subsequent analysis. Iterative operations combined with image fusion algorithms enable rapid, continuous, and accurate measurement of arrayed printheads, aiding in feature recognition. Image recognition methods derive droplet characteristics for precise analysis, and repeated motion stages comprehensively acquire characteristics to generate inkjet characteristics, strongly supporting inkjet system optimization and improving product quality and production efficiency.

This invention discloses a fully automated microscopic scanning and result judgment system for TAP detection. This invention relates to the field of biomedical engineering technology within the bio-industry. The central control and processing unit runs TAP feature analysis logic, which is configured to: identify suspected TAP agglomerates under low magnification using color threshold segmentation; perform multi-focal plane Z-axis tomography on the suspected TAP agglomerate regions under high magnification; generate a clear panoramic image of the TAP agglomerates using a multi-focal plane image fusion algorithm; and extract the morphological and textural features of the agglomerates from the fused image. This invention addresses the three-dimensional stacking characteristics of TAP agglomerates, unlike existing technologies that treat them as defects and discard them. Instead, it creatively introduces Z-axis tomography and multi-focal plane fusion technology, transforming "blurriness" into "features," thus fully acquiring the spatial morphological information of the TAP agglomerates and significantly improving detection sensitivity.

The application provides an infrared and visible light image fusion method and system based on bidirectional semantic-space alignment and differential perception, and relates to the technical field of image processing. In order to solve the problem that the expression capacity is insufficient when the existing method processes complex cross-modal semantic interaction. The technical points of the application include: S1, collecting infrared images and visible light images; S2, fusing infrared and visible light images based on a semantic fusion network, wherein the semantic fusion network is based on a PSFusion network model, the SDFM module of the PSFusion network model is replaced by a differential perception complementary module DACM, the PSFM module is replaced by a deep semantic fusion block DSFB, and a space semantic alignment module SSAM is added after the deep semantic fusion block and the differential perception complementary module; S3, image reconstruction and supervision are performed on the aligned features. The method provided by the application shows superior potential compared with the current advanced image fusion algorithm.

The application provides an infrared and visible light image fusion method based on second-order attention mixed features, first-order and second-order statistics of an infrared image and a visible light image are input, global features and local features of the infrared image and the visible light image are adaptively integrated, so that the fusion image obtains stronger feature expression, meanwhile, pixel intensity and texture details are retained, comparable contrast is maintained while more realistic edges are realized, the output result retains rich edge and texture information transmitted by the input image, has fewer artifacts, and better meets human visual perception; compared with other infrared and visible light image fusion algorithms based on deep learning, the application has better fusion effect.

The application belongs to the technical field of image fusion, and proposes an image fusion controllable modal modulation method based on non-classifier guidance: firstly, a double-branch neural network containing an unconditional reconstruction branch and a conditional fusion branch is constructed, the unconditional branch is used for high-quality fidelity reconstruction of the source image to suppress pre-background noise in the absence of conditional modal guidance, and the conditional branch is used for extracting features from the source image and the conditional modal image, and performing dynamic modulation and fusion in a multi-scale space; secondly, a three-stage joint training strategy is used to train the constructed double-branch neural network; then, in the inference stage, the source image, the conditional modal image and the set guide scale factor are received, unconditional prediction and conditional prediction are performed respectively, and continuous controllable modal injection without retraining is realized through a linear extrapolation formula, and the final fused image is output. The application effectively improves the precision and flexibility of the existing image fusion algorithm in the integration of heterogeneous modal features.

The application discloses a kind of underground coal flow image enhancement and large lump material intelligent identification early warning system, and the application relates to the technical field of coal mine underground coal flow monitoring, by multi-view angle collection coal flow 2D image, vibration-speed coupling offset compensation is carried out in combination with working condition parameter, after 3D profile is extracted actual size by multi-view angle fusion reconstruction, after comparison with dynamic safety threshold, hierarchical early warning is generated and equipment is linked, the advantages of the application are that: by multi-view angle collection underground coal flow 2D image of different angles, vibration-speed coupling offset compensation is carried out to image in combination with working condition parameter, then based on multi-view angle image fusion algorithm, material real 3D profile is reconstructed, actual size of material is accurately extracted and compared with dynamically configured safety threshold, hierarchical early warning signal is generated and conveyor control system is linked to execute targeted disposal, solve the problem that existing technology depends on 2D image projection size judgment, and large lump material is missed due to coal flow stacking shielding.

This invention relates to the field of digital landscape design technology, and discloses a method and system for simulating landscape plant configuration images and visualizing seasonal changes. The method includes: acquiring site image data and performing environmental element identification and light condition analysis; retrieving plant model data and seasonal morphological parameters from a plant model library; fusing plant rendering images with site images using a light-adaptive image fusion algorithm; performing seasonal gradient rendering of plants based on HSV color space mapping to generate seasonal landscape effect images; and calculating long-term plant sizes based on a growth prediction model to generate long-term landscape effect images and visualization reports.

The invention provides a material de-weight statistical method in distribution network project acceptance, and aims to provide a de-weight statistical method in combination with a material volume, and design an algorithm for de-weight through the spatial position and the material volume of a material, and the algorithm can carry out de-weight calculation according to the spatial position and the material volume of the material from a plurality of pictures shot from different angles. The positions of the same material are accurately matched, repeated calculation is avoided, and efficient and accurate material statistics and acceptance are achieved. The second purpose of the invention is to provide a multi-view image fusion algorithm. According to the algorithm, tower pictures shot from different angles can be fused, and a more complete and consistent view is constructed through feature matching and image splicing technologies. The problem of shielding possibly existing under a single view angle is solved, the specific position of each material can be recognized more accurately, and therefore repeated counting in statistics is avoided.

The application discloses a medical image fusion method and system based on cross-modal Mamba and space-frequency cooperation, which comprises an adversarial learning network of a generator and two discriminators, wherein the generator is generally a double-branch encoder-single-branch decoder structure; the fusion image obtained by the application can better retain the clear texture information of the MRI image and fully retain the color information of the PET / SPECT image; the objective evaluation results show that the algorithm is better than the average value of the comparative method by about 24.35%, 3.0%, 14.86%, 24.84%, 21.31%, 5.2% and 21.23% in the seven indexes of spatial frequency, pixel feature mutual information, visual information fidelity, average gradient, edge retention, structural similarity index and wavelet feature mutual information, which further shows that the method can effectively retain the color information of the PET / SPECT image and better fuse the texture information of the MRI image, thereby improving the performance of the existing PET / SPECT and MRI image fusion algorithm.