Accurate classification of weed species in crop plants plays a crucial role in precision agriculture by enabling targeted treatment. Recent studies show that artificial intelligence deep learning (DL) models achieve p...
详细信息
Accurate classification of weed species in crop plants plays a crucial role in precision agriculture by enabling targeted treatment. Recent studies show that artificial intelligence deep learning (DL) models achieve promising solutions. However, several challenging issues, such as lack of adequate training data, inter-class similarity between weed species and intra-class dissimilarity between the images of the same weed species at different growth stages or for other reasons (e.g., variations in lighting conditions, image capturing mechanism, agricultural field environments) limit their performance. In this research, we propose an image based weed classification pipeline where a patch of the image is considered at a time to improve the performance. We first enhance the images using generative adversarial networks. The enhanced images are divided into overlapping patches, a subset of which are used for training the DL models. For selecting the most informative patches, we use the variance of Laplacian and the mean frequency of Fast Fourier Transforms. At test time, the model's outputs are fused using a weighted majority voting technique to infer the class label of an image. The proposed pipeline was evaluated using 10 state-of-the-art DL models on four publicly available crop weed datasets: DeepWeeds, Cotton weed, Corn weed, and Cotton Tomato weed. Our pipeline achieved significant performance improvements on all four datasets. DenseNet201 achieved the top performance with F1 scores of 98.49%, 99.83% and 100% on Deepweeds, Corn weed and Cotton Tomato weed datasets, respectively. The highest F1 score on the Cotton weed dataset was 98.96%, obtained by InceptionResNetV2. Moreover, the proposed pipeline addressed the issues of intra-class dissimilarity and inter-class similarity in the DeepWeeds dataset and more accurately classified the minority weed classes in the Cotton weed dataset. This performance indicates that the proposed pipeline can be used in farming applicat
This study concentrates on consistent object contour extraction method for stereo image segmentation after the object regions in the left image have been obtained. By taking advantage of the epipolar geometry, our app...
详细信息
This study concentrates on consistent object contour extraction method for stereo image segmentation after the object regions in the left image have been obtained. By taking advantage of the epipolar geometry, our approach introduces an energy optimization framework that incorporates both the stereo correspondence term and patch-based object contour probability term. The contour map is generated by integrating the terms of stereo correspondence and patch-based object contour probability;then, the optimal contours are obtained using geodesic distance technology. The core of the proposed method is to build upon an energy optimization framework with two key contributions: first, it incorporates the patch-based object contour probability term that introduces two search strategies to efficiently find the joint nearest neighbor patch pairs for the stereo image pair. The patch-based object contour probability term provides consistent and reliable priors for the contour extraction. Second, previous methods encounter missing pixels in the extracted contour in the occluded regions. Our approach overcomes this limit by introducing the geodesic distance technology to search the optimal contours. Experimental evaluation on Middlebury dataset and Adobe open dataset indicates that the results of our stereo image segmentation are comparable with or of higher quality than state-of-the-art methods. (c) 2018 SPIE and IS&T
There are many patch-based techniques used in image processing, but most of them are heavy to compute with current machines. A dissimilarity measure for images based on patches, inspired from rank distance, called Loc...
详细信息
ISBN:
(纸本)9783642411816;9783642411809
There are many patch-based techniques used in image processing, but most of them are heavy to compute with current machines. A dissimilarity measure for images based on patches, inspired from rank distance, called Local patch Dissimilarity (LPD), was recently introduced. It has very promising results in optical character recognition, but, as other patch-based methods, it is computationally heavy. This work aims at showing that LPD can be improved in terms of efficiency. Several ways of optimizing the LPD algorithm are presented, such as using a hash table to store precomputed patch distances or skipping the comparison of overlapping patches. Another way to avoid the problem of the higher computational time on large sets of images is to turn to local learning methods. Several experiments are conducted on two datasets using both standard machine learning methods and local learning methods. All methods are based on LPD. The obtained results come to support the fact that LPD is a very good dissimilarity measure for images. In this paper, LPD is also used with success for classifying images other than handwritten digits.
暂无评论