Yoshihashi, Ryota, Kawakami, Rei, You, Shaodi, Trinh, Tu Tuan, Iida, Makoto, and Naemura, Takeshi
Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract
Detecting tiny objects in a high-resolution video is challenging because the visual information is little and unreliable. Specifically, the challenge includes very low resolution of the objects, MPEG artifacts due to compression and a large searching area with many hard negatives. Tracking is equally difficult because of the unreliable appearance, and the unreliable motion estimation. Luckily, we found that by combining this two challenging tasks together, there will be mutual benefits. Following the idea, in this paper, we present a neural network model called the Recurrent Correlational Network, where detection and tracking are jointly performed over a multi-frame representation learned through a single, trainable, and end-to-end network. The framework exploits a convolutional long short-term memory network for learning informative appearance changes for detection, while the learned representation is shared in tracking for enhancing its performance. In experiments with datasets containing images of scenes with small flying objects, such as birds and unmanned aerial vehicles, the proposed method yielded consistent improvements in detection performance over deep single-frame detectors and existing motion-based detectors. Furthermore, our network performs as well as state-of-the-art generic object trackers when it was evaluated as a tracker on a bird image dataset. Comment: arXiv admin note: text overlap with arXiv:1709.04666
Yoshihashi, Ryota, Trinh, Tu Tuan, Kawakami, Rei, You, Shaodi, Iida, Makoto, and Naemura, Takeshi
Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract
While generic object detection has achieved large improvements with rich feature hierarchies from deep nets, detecting small objects with poor visual cues remains challenging. Motion cues from multiple frames may be more informative for detecting such hard-to-distinguish objects in each frame. However, how to encode discriminative motion patterns, such as deformations and pose changes that characterize objects, has remained an open question. To learn them and thereby realize small object detection, we present a neural model called the Recurrent Correlational Network, where detection and tracking are jointly performed over a multi-frame representation learned through a single, trainable, and end-to-end network. A convolutional long short-term memory network is utilized for learning informative appearance change for detection, while learned representation is shared in tracking for enhancing its performance. In experiments with datasets containing images of scenes with small flying objects, such as birds and unmanned aerial vehicles, the proposed method yielded consistent improvements in detection performance over deep single-frame detectors and existing motion-based detectors. Furthermore, our network performs as well as state-of-the-art generic object trackers when it was evaluated as a tracker on the bird dataset. Comment: 10 pages, 8 figures
Trinh, Tu Tien, Sullivan, Eric, and Snell, Lane E.
Abstract
The disclosure, in one aspect, provides a method of drilling a wellbore that includes features of drilling the wellbore using a drilling assembly that includes a drill bit that further includes a weight sensor and a torque sensor, determining weight-on-bit using measurements from the weight sensor and torque-on-bit using measurement from the torque sensor during drilling of the wellbore, obtaining measurements for rotational speed of the drill bit and rate of penetration of the drill bit during drilling of the wellbore, determining mechanical specific energy of the bottomhole assembly using the determined weight-on-bit, torque-on-bit and obtained rotational speed of the drill bit and the obtained rate of penetration of the drill bit, and altering a drilling a parameter in response to the determined mechanical specific energy.
Trinh, Tu Tien, Sullivan, Eric, Cheng, Xiaomin C., Prevost, Gregory C., Inanc, Feyzi, Liu, Yi, and Habernal, Jason R.
Abstract
In one aspect, a method of making a drill bit is disclosed that includes selecting a drill bit configuration, obtaining a stress map for the drill bit configuration relating to a drilling operation, performing a mechanical test with an actual drill bit having the selected configuration, and selecting a location on a surface of the drill bit for installing a sensor thereat based on a location of low stress from the stress map and results of the mechanical test, and placing a sensor at the selected location.
Trinh, Tu Tien, Sullivan, Eric C., and Cheng, Xiaomin Chris
Abstract
A cutting element for an earth-boring tool includes an elongated body having a longitudinal axis, a generally planar volume of hard material attached to the elongated body, and a sensor affixed to the elongated body. The sensor may be configured to sense at least one of stress and strain. An earth-boring tool includes a cutting element disposed at least partially within a pocket of a body. Methods of forming cutting elements comprise securing a generally planar volume of hard material to an elongated body, attaching a sensor to the elongated body, and configuring the sensor. Methods of forming earth-boring tools comprise forming a cutting element and securing the cutting element within a recess in a body of the earth-boring tool. Methods of forming wellbores comprise rotating an earth-boring tool comprising a cutting element and measuring at least one of stress and strain.
View/download PDF
