Mango leaf disease segmentation is an essential foundation for accurate disease diagnosis and intelligent grading. The size and shape of mango leaf diseases vary significantly at different times, making it difficult for mainstream semantic segmentation methods to segment disease areas accurately. Therefore, this paper proposes a method called MAU-Net for fine segmentation of mango leaf diseases over the whole period. The MAU-Net is based on the traditional Unet architecture, integrates the Self-Aligning Attention Feature Fusion (SAFF) module and the Multiscale Feature Enhancement (MFE) module, and designs a new loss function DF_Loss. Specifically, the designed SAFF module changes the traditional Unet's skip-connection approach by fusing the global and local two-branch attention mechanisms. It enhances the attention to crucial leaf and disease features at different levels and thus retains richer semantic information about mango leaf diseases. The designed MFE module aims to solve the problem of complex multi-scale disease segmentation in different periods of mango leaves by introducing different scales of cavity convolution to enhance the extraction of disease features at different scales. The designed DF_Loss combines the idea of the similarity measure in Dice Loss and the advantages of the attentional conditioning mechanism in Focal Loss with an additional conditioning factor. It allows the model to focus more on pixels that are difficult to categorize during the learning process, thus improving the segmentation accuracy. MAU-Net achieved 99.21%, 84.33%, 97.1%, and 96.94% of leaf IoU, disease IoU, F1, and mPA metrics on the mango leaf disease dataset. It improved 0.36%, 4.88%, 3.9%, and 1.91% over UNet, and 5.59%, 0.19%, 1.6%, and 2.26% over DeepLabv3+, respectively. Therefore, the present study may provide an accurate method for segmenting mango leaf spots over the whole period and provide a sufficient basis for the accurate analysis of mango leaf diseases.
In today's world, a great amount of people, devices, and sensors are well connected through various online platforms, and the interactions between these entities produce massive amounts of useful information. This process of data production and sharing appears to be on the rise. The growing popularity of this industry, as well as the required development of data sharing tools and technology, pose major threats to an individual's sensitive information privacy. These privacy-related issues may elicit a regularly strong negative reaction and restrain further organizational invention. Researchers have identified the privacy implications of large data collections and contributed to the preservation of data from unauthorised exposure to solve the challenge of information privacy. However, the majority of privacy strategies concentrate solely on traditional data models, such as micro-data. The academe and industry are paying more attention to network data privacy challenges. In this paper, we offer (ℓ, k)-anonymity, a novel privacy paradigm for network data that focuses on maintaining the privacy of both node and link information. Here, original network data will turn to attribute generalization nodes through a complex process, where several algorithms, clustering, node generalization, link generalization and ℓ-diversification will be applied. As a result, (ℓ, k)-anonymous network will be generated and will filter original network data to ensure publishable (ℓ, k)-anonymize data. Hopefully, this anonymity model will have a stronger role against homogeneity attacks of intruders, which will prevent the unauthorized disclosure of sensitive network data for several areas, such as - health sector. This model will also be cost effective and data loss will be controlled using two different ways.
A novel method for inverting a mapping of the multilayer feedforward network is proposed. This method is based upon recursive constrained linear equations constructed by a given desired output, weights, an activation function and unknown inputs. By solving these recursive constrained linear equations by nonlinear and linear programming techniques, some typical inversions from a given output can be derived. Therefore, some new relations can be obtained between the unseen inputs and the given output. Two algorithms based on this method for inverting the mapping of the three-layer feedforward network are derived. The first algorithm is to invert the mapping into a global input space. The second algorithm is developed to invert the mapping into a local input space.< >
This paper presents an edge-preserving fourth order partial differential equation (PDE) for image restoration derived from a new surface-based energy functional. The corresponding fourth order PDE can preserve edges and avoid the staircase effect. The proposed model contains a function of gradient norm as an edge detector, which controls the diffusion speed according to the local structure of the image and preserves more details. Denoising results are given and we have also compared our method with some related PDE models.
Coal mine accident rescue is a huge project. But how to ascertain the place where the accident happened, the scope affected by the accident and rescue route is the key of coal mine's emergency response. In this paper, using the function of GIS's space analysis, taking in the concept of half space, constructing entity of coal mine project - relation model, and designing construction of laneway database, constructing shortest path analysis model for evacuating people and transporting succor materials, simulation system is designed in Visual C++ environment for coal mine's emergency response route. That establishes necessary technology base for improvement the scientific quality and pertinence of preparing coal mine accident rescue plan.
As a significant technology, microarray technology plays an essential role in the life sciences research field. The processing of the microarray image is one of the critical phases of microarray technology. A suitable method provides accurate data for microarray technology and can increase experiment accuracy. Microarray image quality is typically poor due to uneven light, discoloration, temperature, artifacts, and instrument contamination. Segmenting the image becomes a challenging task. Irregular screening and level-setting methods are used to segment the microarray image in this article. First, we use an irregular grid method to divide the microarray image and ensure that each point is in a sub-image. Second, the level set method is used to segment the sub-images. Finally, we review the results and select the appropriate data as the final segmentation result. The result of the experiment shows that the proposed algorithm performs better than some related methods.
Set Technique 3 identification is obtained.The curve evolves driven by the structures of two color polarising images and stops at the region edge of the grains. How to referenceIn order to correctly reference this scholarly work, feel free to copy and paste the following:
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)