Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

8results about How to "Rich texture information" patented technology

Infrared image enhancement method based on polarization information

The invention relates to an image enhancement method, in particular to an infrared image enhancement method based on polarization information. The method comprises the following steps: 1, obtaining polarization images of a target scene at four different angles of an infrared band; 2, acquiring Stokes vector images S0, S1 and S2, a polarization degree image and a polarization angle image; 3, acquiring a temperature image T; 4, acquiring a material image M; 5, obtaining a texture image C; and step 6, combining the material image M, the temperature image T and the texture image C to generate an HSV enhanced image of the target scene so as to realize infrared image enhancement. According to the method, four-angle polarization images of a target scene in an infrared band are obtained through an infrared polarization camera, detection perception of environment target temperature, materials and textures is achieved through polarization information, and then an HSV enhanced image rich in contrast ratio and texture information is obtained through an HSV color space fusion method according to the three-dimensional physical information.
Owner:XIAN INST OF OPTICS & PRECISION MECHANICS CHINESE ACAD OF SCI

A point cloud target detection method based on multimodal fusion network

A point cloud target detection method based on a multimodal fusion network belongs to the technical field of three-dimensional target detection. The present invention improves the point cloud grouping method. The voxel structured features are spliced ​​with the features of the point cloud, which expands the features of the original data, provides more effective information before inputting the network architecture, and expresses each point cloud information more reasonably. The network architecture proposed in the present invention splices and fuses the features of the point cloud and the image. The advantages of the depth information provided by the point cloud data and the rich texture information provided by the image data can be effectively integrated, thereby improving the detection accuracy of three-dimensional objects. It can be applied to autonomous driving, underwater robots and other related fields.
Owner:DALIAN UNIV OF TECH

Indoor robot sensing method based on point cloud multi-sensor fusion mapping

The invention discloses an indoor robot sensing method based on point cloud multi-sensor fusion mapping, and relates to an indoor intelligent robot sensing method, which comprises the following steps of: obtaining multi-frame point cloud images of surrounding environment and obstacles by using a binocular camera and a laser radar; processing and fusing the three-dimensional point cloud images respectively acquired by the two three-dimensional point cloud images; in order to improve the robustness of perception, the three-dimensional point cloud is preprocessed before fusion, and interference of point cloud noise on subsequent matching is avoided; based on the target point cloud data corresponding to each frame of image, the point cloud map is accurately registered, so that the intelligent robot can specifically understand the external environment. According to the method, fusion mapping is carried out based on the point cloud data obtained by different sensors, the environment information of the intelligent equipment is determined, the wrong point cloud determined according to the depth corresponding to each point cloud is filtered out from the target point cloud data, and the mapping effect is optimized by filtering out the wrong point cloud influencing the accuracy of the volume measurement result, so that the accuracy of the volume measurement result is improved. The mapping error is obviously reduced, the sensing precision of the robot is improved, and the robustness of the system is improved.
Owner:SHENYANG INSTITUTE OF CHEMICAL TECHNOLOGY

Construction method of lightweight target detection network adaptive to dark light scene

A construction method of a lightweight target detection network adapted to a dark light scene comprises the steps of construction of a training set, a verification set and a test set and training and verification of a model, and the model training process comprises the steps that S1, an original image F passes through two convolution layers in sequence to eliminate redundant information, and a feature map F1 and a feature map F2 are obtained; s2, the original image F, the feature map F1 and the feature map F2 are input into a feature enhancement network in parallel for semantic information coding, and output feature maps O, O1 and O2 on three paths are obtained; s3, the output feature map O2 and the output feature map O1 are subjected to channel dimension splicing through a layer of up-sampling network, the channel dimension is adjusted through one-time up-sampling, and a spliced convolution feature map is obtained; carrying out element addition on the output feature map O and the spliced convolution feature map, and decoding to obtain an image-enhanced intermediate image; and S4, performing feature extraction on the intermediate image by adopting MobileViT, fusing the extracted feature information of different scales, inputting the fused feature information into a detection head, and predicting a detection result by a network.
Owner:SHANGHAI UNIV +1

A radar echo extrapolation method based on time series motion decomposition and dominant fusion

PendingCN122260263Acorrection biasSuppress false attenuationImage enhancementDigital data information retrievalPattern recognitionNetwork Convergence
This application discloses a radar echo extrapolation method based on temporal motion decomposition and dominant fusion, relating to the field of spatiotemporal sequence prediction technology. The method proposes an innovative network fusion paradigm, specifically including: acquiring implicit long-term / short-term temporal features and explicit global / transient motion features during processing using stacked hierarchical recurrent networks; then, innovatively constructing query vectors using temporal features and key-value vectors using motion features, and fusing them using a heterogeneous cross-attention mechanism to form a fusion architecture "dominated by an implicit state modeling network and referenced by an explicit motion modeling network." Based on this, through a series of collaborative designs such as directional decoupling, reference confidence estimation, and long-term feature constraints on short-term features, noise in long-term / short-term motion features is effectively suppressed, enhancing system stability. This method can improve the problems of blurred texture details and attenuation in high-value echo regions in predicted images.
Owner:天津海洋中心气象台(天津港航气象服务中心) +2

Footprint identification method and device and storage medium

The invention provides a footprint identification method and device and a storage medium, and relates to the technical field of computer vision. The method comprises the following steps: acquiring a footprint image to be identified; performing feature extraction on the to-be-identified footprint image by using a shallow feature extraction model to obtain a shallow image feature corresponding to the to-be-identified footprint image; inputting the shallow image features into a deep feature extraction model to obtain full-foot deep features and K local deep features corresponding to the footprint image to be identified; performing feature fusion on the shallow image feature, the full-foot deep feature and the K local deep features to obtain a fusion feature; and performing footprint identification on the to-be-identified footprint image according to the fusion features. According to the method, feature information of different levels and different parts can be effectively integrated, so that the robustness and accuracy of footprint recognition in a complex environment are improved.
Owner:SOUTH CENTRAL UNIVERSITY FOR NATIONALITIES

Hand model reconstruction method based on feature interaction

ActiveCN116563452Bachieve interactionSolve the problem of low estimation accuracyImage enhancementImage analysisHand reconstructionFeature extraction
The application discloses a hand model reconstruction method based on feature interaction, introduces a processing mode of feature interaction in a hand posture estimation stage, so that hand features and skeleton features can effectively learn from each other, so that in the case of occlusion, the hand gesture recognition result can also have higher accuracy. At the same time, in the aspect of feature extraction, the application of the subnetwork also reduces the parameters of the network and accelerates the convergence speed of the model. Finally, the output of the network is a hand shape containing surface texture information, compared with the traditional hand posture estimation result which only presents a hand shape composed of points and lines, the hand reconstruction model obtained by the application is more in line with the hand shape of a person, and when human-computer interaction is realized, a more real interactive experience can be realized, and the operation accuracy can be improved.
Owner:XIAN UNIV OF TECH

Three-dimensional reconstruction method for crops and crop micro organs in close planting state

PendingCN119131231ASolve reconstruction difficultiesachieve the purpose of rebuildingCharacter and pattern recognition3D modellingFeature extractionThree-dimensional space
The invention discloses a three-dimensional reconstruction method for crops and crop micro organs in a close planting state. The three-dimensional reconstruction method comprises the following steps: step 1, setting a rigid signboard around a target plant or building a micro scene reconstruction environment; step 2, shooting a two-dimensional image; step 3, extracting mixed features of each two-dimensional image by using a feature extraction technology; step 4, performing feature point matching on each two-dimensional image by using a matching technology, and generating a pairing group; 5, estimating the pose of each two-dimensional image by using a motion recovery structure technology; and step 6, projecting the two-dimensional pixels into a three-dimensional space by using a multi-view reconstruction technology to generate a three-dimensional model. The method can effectively solve the difficulties of rice reconstruction and one-time reconstruction of small and micro organs in a close planting state, supports the capture of the characteristics of plants in the close planting state, and can realize one-time reconstruction of the small and micro organs of the plants.
Owner:NANJING AGRICULTURAL UNIVERSITY