Matches in SemOpenAlex for { <https://semopenalex.org/work/W4284888598> ?p ?o ?g. }
- W4284888598 endingPage "3275" @default.
- W4284888598 startingPage "3275" @default.
- W4284888598 abstract "Road crack segmentation based on high-resolution images is an important task in road service maintenance. The undamaged road surface area is much larger than the damaged area on a highway. This imbalanced situation yields poor road crack segmentation performance for convolutional neural networks. In this paper, we first evaluate the mainstream convolutional neural network structure in the road crack segmentation task. Second, inspired by the second law of thermodynamics, an improved method called a recurrent adaptive network for a pixelwise road crack segmentation task is proposed to solve the extreme imbalance between positive and negative samples. We achieved a flow between precision and recall, similar to the conduction of temperature repetition. During the training process, the recurrent adaptive network (1) dynamically evaluates the degree of imbalance, (2) determines the positive and negative sampling rates, and (3) adjusts the loss weights of positive and negative features. By following these steps, we established a channel between precision and recall and kept them balanced as they flow to each other. A dataset of high-resolution road crack images with annotations (named HRRC) was built from a real road inspection scene. The images in HRRC were collected on a mobile vehicle measurement platform by high-resolution industrial cameras and were carefully labeled at the pixel level. Therefore, this dataset has sufficient data complexity to objectively evaluate the real performance of convolutional neural networks in highway patrol scenes. Our main contribution is a new method of solving the data imbalance problem, and the method of guiding model training by analyzing precision and recall is experimentally demonstrated to be effective. The recurrent adaptive network achieves state-of-the-art performance on this dataset." @default.
- W4284888598 created "2022-07-09" @default.
- W4284888598 creator A5006710247 @default.
- W4284888598 creator A5012440924 @default.
- W4284888598 creator A5021485937 @default.
- W4284888598 creator A5046867258 @default.
- W4284888598 creator A5068200433 @default.
- W4284888598 creator A5078842042 @default.
- W4284888598 date "2022-07-07" @default.
- W4284888598 modified "2023-10-13" @default.
- W4284888598 title "A Recurrent Adaptive Network: Balanced Learning for Road Crack Segmentation with High-Resolution Images" @default.
- W4284888598 cites W1849277567 @default.
- W4284888598 cites W1861492603 @default.
- W4284888598 cites W1901129140 @default.
- W4284888598 cites W1963857321 @default.
- W4284888598 cites W2110764733 @default.
- W4284888598 cites W2125188192 @default.
- W4284888598 cites W2194775991 @default.
- W4284888598 cites W2395611524 @default.
- W4284888598 cites W2407692387 @default.
- W4284888598 cites W2560023338 @default.
- W4284888598 cites W2565639579 @default.
- W4284888598 cites W2884561390 @default.
- W4284888598 cites W2891534395 @default.
- W4284888598 cites W2912350898 @default.
- W4284888598 cites W2916798096 @default.
- W4284888598 cites W2924254571 @default.
- W4284888598 cites W2962721361 @default.
- W4284888598 cites W2962914239 @default.
- W4284888598 cites W2964309882 @default.
- W4284888598 cites W2994637710 @default.
- W4284888598 cites W2996290406 @default.
- W4284888598 cites W3022492425 @default.
- W4284888598 cites W3041993346 @default.
- W4284888598 cites W3042932150 @default.
- W4284888598 cites W3083292218 @default.
- W4284888598 cites W3088228871 @default.
- W4284888598 cites W3096030187 @default.
- W4284888598 cites W3159506105 @default.
- W4284888598 cites W3167386507 @default.
- W4284888598 cites W3168995468 @default.
- W4284888598 cites W3190626583 @default.
- W4284888598 cites W3206337481 @default.
- W4284888598 cites W3217288639 @default.
- W4284888598 cites W4200218075 @default.
- W4284888598 cites W4200551223 @default.
- W4284888598 cites W4205169960 @default.
- W4284888598 cites W4288325606 @default.
- W4284888598 cites W4298289240 @default.
- W4284888598 doi "https://doi.org/10.3390/rs14143275" @default.
- W4284888598 hasPublicationYear "2022" @default.
- W4284888598 type Work @default.
- W4284888598 citedByCount "5" @default.
- W4284888598 countsByYear W42848885982023 @default.
- W4284888598 crossrefType "journal-article" @default.
- W4284888598 hasAuthorship W4284888598A5006710247 @default.
- W4284888598 hasAuthorship W4284888598A5012440924 @default.
- W4284888598 hasAuthorship W4284888598A5021485937 @default.
- W4284888598 hasAuthorship W4284888598A5046867258 @default.
- W4284888598 hasAuthorship W4284888598A5068200433 @default.
- W4284888598 hasAuthorship W4284888598A5078842042 @default.
- W4284888598 hasBestOaLocation W42848885981 @default.
- W4284888598 hasConcept C108583219 @default.
- W4284888598 hasConcept C111919701 @default.
- W4284888598 hasConcept C127162648 @default.
- W4284888598 hasConcept C154945302 @default.
- W4284888598 hasConcept C162324750 @default.
- W4284888598 hasConcept C187736073 @default.
- W4284888598 hasConcept C2780451532 @default.
- W4284888598 hasConcept C31972630 @default.
- W4284888598 hasConcept C41008148 @default.
- W4284888598 hasConcept C76155785 @default.
- W4284888598 hasConcept C81363708 @default.
- W4284888598 hasConcept C89600930 @default.
- W4284888598 hasConcept C98045186 @default.
- W4284888598 hasConceptScore W4284888598C108583219 @default.
- W4284888598 hasConceptScore W4284888598C111919701 @default.
- W4284888598 hasConceptScore W4284888598C127162648 @default.
- W4284888598 hasConceptScore W4284888598C154945302 @default.
- W4284888598 hasConceptScore W4284888598C162324750 @default.
- W4284888598 hasConceptScore W4284888598C187736073 @default.
- W4284888598 hasConceptScore W4284888598C2780451532 @default.
- W4284888598 hasConceptScore W4284888598C31972630 @default.
- W4284888598 hasConceptScore W4284888598C41008148 @default.
- W4284888598 hasConceptScore W4284888598C76155785 @default.
- W4284888598 hasConceptScore W4284888598C81363708 @default.
- W4284888598 hasConceptScore W4284888598C89600930 @default.
- W4284888598 hasConceptScore W4284888598C98045186 @default.
- W4284888598 hasFunder F4320321001 @default.
- W4284888598 hasFunder F4320324174 @default.
- W4284888598 hasFunder F4320326832 @default.
- W4284888598 hasFunder F4320335777 @default.
- W4284888598 hasIssue "14" @default.
- W4284888598 hasLocation W42848885981 @default.
- W4284888598 hasOpenAccess W4284888598 @default.
- W4284888598 hasPrimaryLocation W42848885981 @default.
- W4284888598 hasRelatedWork W2005437358 @default.
- W4284888598 hasRelatedWork W2517104666 @default.