Object tracking is one of the most important components in numerous applications of computer vision. Remote sensing videos provided by commercial satellites make it possible to extend this topic into the earth observation domain. In satellite videos, typical moving targets like vehicles and planes only cover a small area of pixels, and they could easily be confused with surrounding complex ground scenes. Similar objects nearby in satellite videos can hardly be differed by appearance details due to the resolution constraint. Thus, tracking drift caused by distractions is also a thorny problem. Facing challenges, traditional tracking methods such as correlation filters with hand-crafted visual features achieve unsatisfactory results in satellite videos. Methods based on deep neural networks have demonstrated their superiority in various ordinary visual tracking benchmarks, but their results on satellite videos remain unexplored. In this article, deep learning technologies are applied to object tracking in satellite videos for better performance. A simple regression network is used to combine a regression model with convolutional layers and a gradient descent algorithm. The regression network fully exploits the abundant background context to learn a robust tracker. Instead of handcrafted features, both appearance features and motion features, which are extracted by pretrained deep neural networks, are used for accurate object tracking. In cases when the tracker encounters ambiguous appearance information, the motion features could provide complementary and discriminative information to improve tracking performances. Experimental results on various satellite videos show that the proposed method achieves better tracking performance than other state-of-the-arts.
A colon delivery system has been used to improve the bioavailability of glycyrrhizin, a glycoside of glycyrrhetic acid. The bioavailability of glycyrrhizin is low when administered in conventional oral galenic dosage forms because glycyrrhizin is enzymatically hydrolysed both in the stomach and in the intestine. It was reasoned that if large amounts of glycyrrhizin were directly delivered to the colon, enzymatic activity should be reduced due to saturation so that intact glycyrrhizin could be absorbed into the systemic circulation. Based on this assumption, pressure-controlled colon delivery capsules (PCDCs) were used as a colon delivery system. Eight types of glycyrrhizin solution were prepared and were introduced into PCDCs. After oral administration of the test PCDCs to beagle dogs, blood samples were obtained over 24 h and plasma glycyrrhizin concentrations were measured by an HPLC method. With PCDCs containing aqueous glycyrrhizin and propylene glycol solutions, plasma glycyrrhizin levels were extremely low and the bioavailabilities of glycyrrhizin were 0.6% and 0.4%, respectively. When Labrasol was added to both types of glycyrrhizin solution, the bioavailability was improved to 4.6% for aqueous solution and 3.8% for propylene glycol solution. When a surfactant, Polysorbate 80, was added in combination with Labrasol, synergistic effects were not obtained. Furthermore, dose-dependent effects of Polysorbate 80 were not obtained. Labrasol, which is a component of self-emulsifying drug delivery systems (SEDDS), has been shown to strongly improve the bioavailability of glycyrrhizin from the colon.
A new pH-sensitive polymer, P-4135F, was evaluated as a colon delivery device for norfloxacine (NFLX) which is used for the therapy of patients with Vero toxin-producing Escherichia coli gastroenteritis. P-4135F has a dissolution threshold pH of 7.2 which is higher than the conventional pH-sensitive polymers, Eudragit S100 and L100. To compare the dissolution site of P-4135F coated tablets with other enteric polymer coatings, mini-tablets containing sodium fluorescein (FL) as a model drug were prepared by coating them with the three polymers. After oral administration of FL mini-tablets to rats, the first-appearance time, Ti, of FL into the systemic circulation was measured. The Tis were 0.7+/-0.2 h for Eudragit L100, 1.8+/-0.4 h for S100 and 2.0+/-0.3 h for P-4135F. Direct inspection of the dissolution process of the FL mini-tablets after oral administration to rats was performed by abdominal incision studies. All of the coated FL mini-tablets started to dissolve in the rat ileum. The dissolution sites were identified to be proximal to the ileocecal junction for P-4135F, at the middle part of the ileum for Eudragit S100 and at the proximal part of the ileum for Eudragit L100. NFLX tablets with different membrane thicknesses of P-4135F were prepared and were orally administered to beagle dogs. The colon delivery efficiency was evaluated by measuring the Ti of NFLX into the systemic circulation. The mean Tis were 1.33+/-0.33 h for 56.8+/-0.5 microm membranes, 3.75+/-0.25 h for 64.6+/-0.7 microm membranes, 4.00+/-1.00 h for 70.5+/-0.5 microm membranes and 3.00+/-1.00 h for 74.9+/-0.4 microm membranes. By comparing the Ti, 4.33+/-0.33 h, obtained after oral administration of NFLX in a pressure-controlled colon delivery capsule, and the colon arrival time, 3.5+/-0.3 h, determined by a sulfasalazine test in beagle dogs. P-4135F coated NFLX tablets appeared to dissolve and disintegrate before reaching the colon. Studies using rats and beagle dogs have suggested that P-4135F dissolves in the lower part of the small intestine, i.e., the ileum. These studies also suggest that this new polymer will be useful for the delivery of NFLX to the lower part of the small intestine.
Large quantities of pressure-controlled colon delivery capsules (PCDCs) were prepared by a Hicoater-mini pharmaceutical coating machine and colon delivery efficiencies were evaluated in man. Caffeine powder as a model drug was suspended with a polyethylene glycol (PEG) 1000 suppository base at 50 degrees C, and was hardened in no. 0- and no. 2-sized capsular shapes. The capsule-shaped suppositories were coated with 5% w/v ethanolic ethylcellulose (7G grade) solution using the coating machine. By increasing the coating weight of ethylcellulose from 28.6 +/- 1.1 mg to 45.3 +/- 0.2 mg, the mean coating thickness of no. 0 PCDCs increased from 56 +/- 1 microm to 64 +/- 1 microm. With no. 2 PCDCs, the mean coating thickness increased from 50 micro +/- 1 microm to 57 +/- 1 microm by increasing the coating weight of ethylcellulose from 8.1 +/- 0.5 mg to 11.2 +/- 0.3 mg. The no. 0 PCDCs, having a mean ethylcellulose coating membrane thicknesses of 56 +/- 1 microm (type 1) and 64 +/- 1 microm (type 2), as well as no. 2 PCDCs, having thicknesses of 50 +/- 1 microm (type 3) and 57 +/- 1 microm (type 4), were used for in-vivo evaluation in man. After oral administration of test preparations containing 75 mg of caffeine, saliva samples were obtained and salivary caffeine levels were measured by an HPLC method. The first appearance time, Ti, of caffeine in the saliva was used as a parameter for the estimation of the release time of caffeine from PCDCs in the gastrointestinal tract. The mean Ti values of no. 0 PCDCs were 3.3 +/- 0.3 h for type- 1 and 5.3 +/- 0.3 h for type-2 preparations while the mean Ti values of no. 2 PCDCs were 4.3 +/- 0.5 h for type 3 and 5.3 +/- 0.3 h for type 4. There were good correlations between ethylcellulose coating membrane thicknesses and in-vivo Ti values. A colon arrival time of 5 h was reported in our subjects by gastrointestinal magnetomarkergraphy. PCDCs having a mean coating thickness of 64 +/- 1 microm for no. 0 capsules and of 57 +/- 1 microm for no. 2 capsules were thought to deliver caffeine to the human colon efficiently.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.