Online shearer-onboard personnel detection method for the intelligent fully mechanized mining face

As unmanned coal mining technology gradually replaces the traditional mechanized coal mining technology, shearer operation mode is changed from local control to remote control in intelligent fully mechanized mining face. In remote control mode, it is difficult to protect the personnel who straying into the shearer operator space without observation and reminder from the shearer operator. Hence, it is necessary to establish an intelligent personnel detection method to protect the safety of coal miners in intelligent unmanned mining face. The environment of low and non-uniform illumination in fully mechanized coal mining face has seriously restricted the application of personnel detection technology based on visible light imaging. Therefore, a personnel detection method based on infrared thermal imaging is proposed in this paper to solve the disadvantages of using visible light imaging in downhole applications. On this basis, a spatiotemporal guided filter is proposed to harmonize the relationship between edge-preserving and noise-removing. Then, an improved Lucas-Kanade method based on the adaptive-size window is utilized to achieve a more robust personnel detection. Moreover, the personnel detection in the shearer operating space is realized based on epipolar geometry and morphology processing. Finally, the laboratory experiment and industrial test are carried out to evaluate the proposed method, and the results indicate the feasibility and superiority of the proposed methods which show considerable application prospects.

Pyrrole Hemithioindigo Antimitotics with Near-Quantitative Bidirectional Photoswitching Photocontrol Cellular Microtubule Dynamics with Single-Cell Precision

<div> <div> <div> <p>Photoswitchably bioactive reagents, known as “photopharmaceuticals”, promise powerful applications in high-precision biological research. Yet most photoswitch scaffolds cannot be quantitatively bidirectionally photoisomerised, so they suffer from residual background activity that can confound experiments. We rationally designed photopharmaceuticals using the emerging near-quantitative photoswitch pyrrole hemithioindigo (<b>PHTubs</b>), to isomer-specifically inhibit the cytoskeletal protein tubulin. These <b>PHTub</b> reagents allow simultaneous visible-light imaging and photoswitching in live cells, where they could be used for cell-precise photomodulation of microtubule dynamics, and photocontrol over cell cycle progression and cell death. This is, as far as we know, the first use of a hemithioindigo photopharmaceutical for high-spatiotemporal-resolution biological control in live cells. This work opens up new horizons for high-precision microtubule research using <b>PHTubs</b>; and shows the cellular applicability of the near-quantitative photoswitch pyrrole hemithioindigo as a valuable scaffold for photocontrol of a range of other biological targets. </p> </div> </div> </div>

Pyrrole Hemithioindigo Antimitotics with Near-Quantitative Bidirectional Photoswitching Photocontrol Cellular Microtubule Dynamics with Single-Cell Precision

<div> <div> <div> <p>Photoswitchably bioactive reagents, known as “photopharmaceuticals”, promise powerful applications in high-precision biological research. Yet most photoswitch scaffolds cannot be quantitatively bidirectionally photoisomerised, so they suffer from residual background activity that can confound experiments. We rationally designed photopharmaceuticals using the emerging near-quantitative photoswitch pyrrole hemithioindigo (<b>PHTubs</b>), to isomer-specifically inhibit the cytoskeletal protein tubulin. These <b>PHTub</b> reagents allow simultaneous visible-light imaging and photoswitching in live cells, where they could be used for cell-precise photomodulation of microtubule dynamics, and photocontrol over cell cycle progression and cell death. This is, as far as we know, the first use of a hemithioindigo photopharmaceutical for high-spatiotemporal-resolution biological control in live cells. This work opens up new horizons for high-precision microtubule research using <b>PHTubs</b>; and shows the cellular applicability of the near-quantitative photoswitch pyrrole hemithioindigo as a valuable scaffold for photocontrol of a range of other biological targets. </p> </div> </div> </div>

Internal defect scanning of sweetpotatoes using interactance spectroscopy

While standard visible-light imaging offers a fast and inexpensive means of quality analysis of horticultural products, it is generally limited to measuring superficial (surface) defects. Using light at longer (near-infrared) or shorter (X-ray) wavelengths enables the detection of superficial tissue bruising and density defects, respectively; however, it does not enable the optical absorption and scattering properties of sub-dermal tissue to be quantified. This paper applies visible and near-infrared interactance spectroscopy to detect internal necrosis in sweetpotatoes and develops a Zemax scattering simulation that models the measured optical signatures for both healthy and necrotic tissue. This study demonstrates that interactance spectroscopy can detect the unique near-infrared optical signatures of necrotic tissues in sweetpotatoes down to a depth of approximately 5±0.5 mm. We anticipate that light scattering measurement methods will represent a significant improvement over the current destructive analysis methods used to assay for internal defects in sweetpotatoes.

Seeing Pedestrian in the Dark via Multi-Task Feature Fusing-Sharing Learning for Imaging Sensors

Pedestrian detection is an essential problem of computer vision, which has achieved tremendous success under controllable conditions using visible light imaging sensors in recent years. However, most of them do not consider low-light environments which are very common in real-world applications. In this paper, we propose a novel pedestrian detection algorithm using multi-task learning to address this challenge in low-light environments. Specifically, the proposed multi-task learning method is different from the most commonly used multi-task learning method—the parameter sharing mechanism—in deep learning. We design a novel multi-task learning method with feature-level fusion and a sharing mechanism. The proposed approach contains three parts: an image relighting subnetwork, a pedestrian detection subnetwork, and a feature-level multi-task fusion learning module. The image relighting subnetwork adjusts the low-light image quality for detection, the pedestrian detection subnetwork learns enhanced features for prediction, and the feature-level multi-task fusion learning module fuses and shares features among component networks for boosting image relighting and detection performance simultaneously. Experimental results show that the proposed approach consistently and significantly improves the performance of pedestrian detection on low-light images obtained by visible light imaging sensor.

Highly Standardized Rotational Photography of the Torso

Over the last several decades, medical imaging technologies have proven so anatomically insightful that they have all but taken over management in many clinical situations. MRI, CT, and PET scanning technologies employ stringently controlled rotational data-harvest protocols. By contrast, standardization of visible-light imaging of human anatomy has lagged dramatically. This article is the first to present a rigidly standardized rotational protocol to photographically record human surface anatomy and permit subsequent analysis with less than 2% image variance.