Camera Shooting Research Articles

EVALUATION OF THE DYNAMIC CHARACTERISTICS OF THE TRANSITION OF A STOP TO A STAND IN BASKETBALL AS THE BASIC BASIS OF A PLAYER'S TECHNICAL ACTIONS

The article reveals the peculiarity of the mechanism of performing basic technical actions in basketball, followed by the adoption of dynamic equilibrium after stopping and maintaining postural stabilization in the basketball player’s stance. The results of the study. A method for assessing dynamic characteristics using video analysis of performing a stop with two steps and a jump and moving into a basketball player’s stance. After the calibration procedure of the plane of the sup-port zone in the Kinovea program, the length and width of the support area of the subjects were determined on the image of the video fi le using the external extreme points of the subjects’ feet, as a result of which it became possible to fi x the Spopping — Type of stop, i. e. the type of stop: jump (jump stop); right (stop under the right foot); left (stop under the left leg). the indicators aff ecting the qualitative characteristics of the transition to the rack after stopping the changes in the links of the musculoskeletal system in degrees were also evaluated. A number of indicators were recorded, including the time characteristics of the motor action before stopping, including a number of parameters with subsequent calcula-tion of the time of passage of a 6-meter section before stopping in seconds, the time of steps before stopping, the time of steps directly during stopping and the length of steps with the determination of the duration of dynamic stability — fi nd-ing balance after movement and stopping with an assessment the length and width of the support area, followed by the calculation of the support area. The time of dynamic stabilization was determined from the initial moment of support of the two legs of the subject during the stop to the moment of absence of movement of the trunk after the stop. The error in determining the time characteristics of the study was 0.02 seconds at a camera shooting frequency of 50 Hz. After the subjects completed all three types of stops, the video recording was stopped and marked for each subject. The height of the location of the video camera above the level of the. The analysis of video fi les to obtain the time characteristics of three types of stops in basketball was carried out on a computer running the Windows 10 operating system (64 bit) using a free MPC-HC (64bit) audio and video fi le player version 1.7.9 (846eff 0). The measurement of the linear characteristics of the support area of the subject after stopping was carried out using the Kinovea program version 0.9.5. It performed the functions of capturing, slowing down, comparing, commenting and measuring movement in a video file.

A. 볼로딘의 희곡 다섯 밤의 무대화와 영화화 연구

The purpose of this paper is to study how A. Volodin's play Five Nights has been reproduced and recreated through the artistic media of theater and film. And the purpose is to study how the director (film director) discovered and supplemented the hidden side of the play. In other words, the purpose of this paper is to examine how a play, which is a written text, is converted into a form through a play delivered through words and actions, and a movie delivered through video through camera shooting, and the aspects and meaning of that transformation. To this end, this paper selected Topstanogov's play from Vedette in 1959 and Mikhalkov's film from 1978, and studied the directors' original interpretations and complementary aspects of the original text in different time periods and genre characteristics. This paper proposes that it can be actively utilized in the research methodology of media conversion of works of art, which is emerging as a trend in modern culture and arts research.

Spatial infiltration and redistribution of light crude oil in heterogeneous water-bearing soil layers under different hydrogeological processes.

Two physical models were used to simulate the infiltration and redistribution process of light crude oil after leakage in a heterogeneous soil layer following water level variation and rainfall. Migration fronts and redistribution characteristics of oil during gravity seepage, water level variation, and rainfall were obtained using charge-coupled device (CCD) camera shooting and cyan-magenta-yellow‒black (CMYK)-based gray analysis, which were employed efficiently and at a low cost. Then, the influencing factors and migration mechanisms were examined. Finally, the soil water and oil contents were measured to verify the simulation results. The results are as follows: (1) the geologic lens and fine-coarse interface can intercept oil, resulting in a local highly contaminated area. (2) The crude oil infiltration path and velocity varied greatly with the different soil types and initial water contents. Within a certain range, the higher the initial water content is, the higher the lateral and vertical infiltration speeds. (3) The oil redistribution process was dominated by vertical infiltration under the condition of water level variation or rainfall, but oil-water displacement and the capillary pressure caused some oil to move horizontally near the geologic lens and fine-coarse interface. (4) Water level variation resulted in a synchronous rise or fall of the oil accumulation area, but rainfall caused it to move up. (5) Water level variation and rainfall imposed a certain influence on the periodic accumulation and release of crude oil in heterogeneous soil, especially in the presence of geologic lenses and lithologic interfaces.

Robust Watermarking Against Camera Shooting for PowerPoint Presentation

Robust Watermarking Against Camera Shooting for PowerPoint Presentation

Application of 3D Printing Technology in Customized Garment Production

Abstract In this paper, human body images are acquired by DSLR camera shooting, and the photos are grayscaled based on the HIS model. Then, the different operations weaken the similar parts between the images, and the other parts of the photos are highlighted. The image is separated from the background by image binarization. After the contour detection of the image, the pixel coordinates of the contour points in the image are deposited into a dynamic array by the size judgment of the pixel values. The subsequent size calculation is called to make a pad. According to the human body, clothing size automation measurement has been derived from the human body characteristics of the size, the use of three-dimensional printing technology to produce clothing to meet consumer demand, and three-dimensional printing technology to support the production of customized clothing under the experimental analysis. It is verified that the data difference between the actual size of the subject’s ready-made clothes and the size of the clothes produced based on 3D printing technology is kept within ±0.48cm. The virtual fitting pressure is less than the comfort threshold of women’s upper body pressure (2.75kpa), which indicates that the clothes produced by this research will not cause physiological discomfort to the human body and, at the same time, meet the consumer’s personalized clothing customization needs.

Single image defogging via multi-exposure image fusion and detail enhancement

Outdoor cameras play an important role in monitoring security and social governance. As a common weather phenomenon, haze can easily affect the quality of camera shooting, resulting in loss and distortion of image details. This paper proposes an improved multi-exposure image fusion defogging technique based on the artificial multi-exposure image fusion (AMEF) algorithm. First, the foggy image is adaptively exposed, and the fused image is subsequently obtained via multiple exposures. The fusion weight is determined by the saturation, contrast, and brightness. Finally, the image fused by a multi-scale Laplacian algorithm is enhanced with simple adaptive details to obtain a clearer defogging image. It is subjectively and objectively verified that this algorithm can obtain more image details and distinct picture colors without a priori information, effectively improving the defogging ability.

SPECIAL EFFECTS USED IN CREATING 3D ANIMATED SCENES PART II

Today, the computers have made special effects so real, very hard to distinguish from the real elements on the screen, that we can barely notice them. The 3D animation technology development mirrors the progress of technology in the 20th century from mechanization and industrial to computerization and virtual. While the special 3D animation from today could not exist without computers, people are still the ones who conceive these effects. There are many opinions according to which 3D animation will have a much greater impact than it has had up to now. Filming animated 3D scenes offers various opportunities for achieving direct camera shooting. 3D Graphics is a lot more complex, it works with textures, volumes, light schemes and everything is being realized with the help of the computer and of the intelligent mind of the 3D graphics editor. Also, the computer-aided design, the integrated machine-tool systems with automatic control, industrial robots, etc., are some of the current mutations of industrial production that strongly influence product design. The 3D programs are spectacular; thanks to the developments from the hard and software technology, lighting and dynamic renderings specific to 3D scenes can be integrated more and more easily even in complex scenes. The viewer will not even notice that some of the nature landscapes will be generated by the 3D graphic editor. The aim of this article is to briefly present the geometric modelling of some component elements of the proposed 3D scene. Like any other developed product, it has to replace one that already exists, and for it to be accomplished an essential criterion is to introduce a new user facility or to improve the technical performance of the predecessor. Competition also led to diversity, the producers had to capture unexplored market sectors and bring improvements specific to a certain job. The aesthetic valences are ensured by the use of high-quality plastic, metal and ceramic materials and finishing, but also by an arrangement of different primary or combined geometric shapes that confer the status of a “decor” object, when we look at the design of new products as an art.

Error Analysis and Accuracy Optimization for the Hand-eye Calibration of Line Structured Light

The rapid development of machine vision technology in recent years has resulted in applications in engineering scenarios such as weld tracking, visual measurement, and precision gripping. Focusing on the fundamental steps of applying machine vision technology, the “hand-eye calibration” process is analyzed based on scenarios with a line-structured light camera and calibration ball. We found out that noise and camera pose are the main sources of calibration error. To address these issues, an improved K-medoids noise pre-processing algorithm and a camera shooting attitude control scheme are proposed and applied. The experimental results showed that when the radius of the tangent circle/radius of the calibration sphere approximates 0.618, the root means a square error of the repeatability accuracy of hand-eye calibration could be controlled within 0.23 mm, which fully meets the application requirement. Considering the low cost of line-structured light cameras, the findings of this study contribute to the large-scale diffusion of machine vision technology in engineering practices.

Development and Performance Test of a Simulation-Based Tractor-Attachable Wind-Blast-Type Onion Stem Cutting Machine

Stem cutting process of many steps in onion farming involves the least supply of machines. Machines developed in foreign countries are not suitable owing to the geographic characteristics of Korea; it is necessary to develop customized machines. Therefore, in this study, a tractor-attachable blast-type onion stem cutting machine was developed to address these issues. To determine the installation angle of a detachable blade with a 300-mm length, 115-mm width, 5-mm thickness, and a cutting angle of 45° at the tip, we compared the simulation results with the analysis results obtained by fabricating a simple experimental device. Despite small differences in values, the R2 between the two results was found to be 0.99; if using a regression equation, they could be adequate for use in estimation. Moreover, by measuring the motor power consumption, wind speed, and high-speed camera shooting, the stem was raised, regardless of its water content, when the installation angle was greater than 30°. Motor power consumption could also be reduced when the installation angle was 30° rather than 40°. Therefore, the optimal cutter blade installation angle for the tractor-attachable blast-type onion stem cutting machine was confirmed to be 30°. The developed machine was tested at various rotating speeds. The best performance was obtained when the test was repeated thrice at a driving speed of 0.4 m/s. The remaining onion stem length averaged 9.98 cm, the standard deviation was 4.72 cm and the stem cutting rate was 96.8%.

Development and experimental verification of C-arm camera shooting locator

This study aimed to develop a self-made C-arm camera shooting locator and verify its accuracy and advantages. A total of 60 physicians and nurses from the Surgical System of Sanmen People’s Hospital, Zhejiang Province, China, were randomly selected as filming operators. The C-arm machine with a self-made locator and a C-arm machine without a locator were used to measure the center of the circular plate. The iron nails were used to shoot. The distance between the iron nail and the center point of the circular display area on display was defined as the shooting deviation. When it was less than 3 cm, the shooting was stopped. The number of shots, total shooting time, and first-shot deviation in the C-arm camera shooting groups with and without the locator were statistically analyzed, and the advantages and disadvantages of the two were compared. The average number of shots, average total shooting time, and average first-shot deviation of the C-arm camera using the locator were significantly better than those in the group without the locator, and the differences were statistically significant. When the shooting distance (X) was equal to 30 cm and the shooting angle (Y) was equal to 0°, the average number of shots, average total shooting time, and average first-shot deviation were optimal. The C-arm camera shooting locator can improve the shooting accuracy of the C-arm camera and effectively reduce the number of shots and total shooting time. Hence, it can be applied in clinical and surgical practice.

Citrus green fruit detection via improved feature network extraction.

It is crucial to accurately determine the green fruit stage of citrus and formulate detailed fruit conservation and flower thinning plans to increase the yield of citrus. However, the color of citrus green fruits is similar to the background, which results in poor segmentation accuracy. At present, when deep learning and other technologies are applied in agriculture for crop yield estimation and picking tasks, the accuracy of recognition reaches 88%, and the area enclosed by the PR curve and the coordinate axis reaches 0.95, which basically meets the application requirements.To solve these problems, this study proposes a citrus green fruit detection method that is based on improved Mask-RCNN (Mask-Region Convolutional Neural Network) feature network extraction. First, the backbone networks are able to integrate low, medium and high level features and then perform end-to-end classification. They have excellent feature extraction capability for image classification tasks. Deep and shallow feature fusion is used to fuse the ResNet(Residual network) in the Mask-RCNN network. This strategy involves assembling multiple identical backbones using composite connections between adjacent backbones to form a more powerful backbone. This is helpful for increasing the amount of feature information that is extracted at each stage in the backbone network. Second, in neural networks, the feature map contains the feature information of the image, and the number of channels is positively related to the number of feature maps. The more channels, the more convolutional layers are needed, and the more computation is required, so a combined connection block is introduced to reduce the number of channels and improve the model accuracy. To test the method, a visual image dataset of citrus green fruits is collected and established through multisource channels such as handheld camera shooting and cloud platform acquisition. The performance of the improved citrus green fruit detection technology is compared with those of other detection methods on our dataset. The results show that compared with Mask-RCNN model, the average detection accuracy of the improved Mask-RCNN model is 95.36%, increased by 1.42%, and the area surrounded by precision-recall curve and coordinate axis is 0.9673, increased by 0.3%. This research is meaningful for reducing the effect of the image background on the detection accuracy and can provide a constructive reference for the intelligent production of citrus.

Algorithm Design of Mirror Tracking System

In the current national defense background, the requirements for the performance of weapons are increasing day by day, the speed of the flying target and the flight state provide important parameters for the improvement of the characteristics of weapons, ballistics and ammunition. In the shooting range test, the high-speed flying target is tracked and photographed to obtain important indicators such as the flight status of the flying target in high-speed flight. This paper studies a ballistic tracking camera system based on a rotating mirror. It is composed of a system, which can track the flight trajectory of the flying target in the air in real time, and obtain important information of the flying target in flight through high-speed camera shooting. Design the mirror tracking system to capture the flight status of the flying target, analyze the flight trajectory parameters of the target through MATLAB, and optimize the accuracy and response speed of the motor control algorithm. The motor turns the mirror to reflect the flying target to the high-speed camera. The feasibility of the system is ver ifi ed. Th e tot al sc ann ing range of the system for tracking flying targets is 90°, and the scanning accuracy is ±0.5°.

Automatic imaging system mounted on boom sprayer for crop scouting using an off-the-shelf RGB camera

This article describes an imaging system mounted on the boom of a self-propelled sprayer, which was developed to automatically capture and store high-definition video images, while handling agrochemical spray and machine vibration. This was accomplished using an off-the-shelf camera (DSC-RX0M2, Sony) and two single board computers. To enable automatic operation, the system was designed to start and terminate along with the engine. All components were powered successfully except for the camera which runs only on its own battery. This was overcome by delaying powering certain hardware and software components of the system for different periods to allow consistent initialization of the entire system. Accordingly, the current version of the system starts imaging 15 min after engine start-up and guarantees performance for 3 h of consecutive video capturing. After several field trials, the imaging system became able to save images with the required quality. The developed software is able to handle images at a rate of 30 frames per second without delay. In order to handle dynamic light conditions, the camera shooting mode was set to manual at a very high shutter speed. This provided detailed images at both sunny and cloudy conditions. While the current captured images proved to be adequate to start machine vision system development, there is a setback when trying to get images at early morning or evening as they tend to be too dark. This will be handled while upgrading the system towards a machine vision for the purpose of automatic scouting.

Retracted] Realization of a Noncontact IC Chip with Embedded Ferroelectric Memory in an Auxiliary Timing Device for Sports Games

The research direction of the new generation of embedded memory can be summarized into two types of embedded nonvolatile memory and embedded volatile memory; the research on online testing of embedded memory started in the past ten years, and there are few research results. This article analyzes the feasibility of the noncontact IC chip in the embedded ferroelectric memory of the sports game auxiliary timing device and is aimed at obtaining an optimized embedded ferroelectric memory by analyzing the relevant data to achieve the update and update of the sports game timing device system. Early sports event timing methods generally use manual timing (stopwatch) or camera shooting timing; this method is inefficient, poor real‐time, huge workload, and prone to errors. This research mainly focuses on the analysis and discussion of the material structure and performance of the embedded ferroelectric memory and the process of noncontact IC chip. This article uses custom welding circuit technology to prepare the best ferroelectric filter in the test part and verifies the influence of temperature on the material; in order to understand the properties of ferroelectric materials at the electronic and atomic level, a first‐order statistical method is obtained. The numerical calculation results of the experiment verify that the evaluation value of the serial port synchronization module as a whole exceeds the pulse synchronization; the network synchronization as a whole exceeds the code synchronization, and the result of the network time service module is the opposite, but as a whole, each module of the noncontact IC chip has strong performance adaptability; in the application of auxiliary timing, the maintainability of noncontact IC chip is quite outstanding, and the maximum value is 7.97; a large number of complex simulation system tasks can be completed by simple and direct tasks.

Liquid CO2 Phase-Transition Rock Fracturing: A Novel Technology for Safe Rock Excavation

In order to determine the applicability of liquid CO2 phase-transition fracturing technology in rock mass excavations, the principles of CO2 phase-transition fracturing were analyzed, and field tests of liquid CO2 phase-transition fracturing were performed. An “Unmanned Aerial Vehicle (UAV) camera shooting + Microstructure Image Processing System (MIPS) analyzing” method was used to acquire the rock mass characteristics. Further, the Hilbert–Huang Transform (HHT) energy analysis principle was adopted to analyze the characteristics of fracturing vibration waves. The experimental results showed that during the process of fracturing, there were both dynamic actions of rock breakage due to excitation stress wave impacts, and quasi-static actions of rock breakage caused by gasification expansion wedges. In semi-infinite spaces, rock-breakage zones can mainly be divided into crushing zones, fracture zones, and vibration zones. At the same time, under ideal fracturing effects and large volumes, the fracturing granularity will be in accordance with the fractal laws. For example, the larger the fractal dimensions, the higher the proportion of small fragments, and vice versa. Moreover, the vibration waves of the liquid CO2 phase-transition fracturing have short durations, fast attenuation, and fewer high-frequency components. The dominant frequency band of energy will range between 0 and 20 Hz. The liquid CO2 phase-transition fracturing technology has been observed to overcome the shortcomings of traditional explosive blasting methods and can be applied to a variety of rock types. It is a safe and efficient method for rock-breaking excavations; therefore, the above technology effectively provides a new method for the follow-up of similar engineering practices.

Screen-Shooting Resilient Watermarking Scheme via Learned Invariant Keypoints and QT.

This paper proposes a screen-shooting resilient watermarking scheme via learned invariant keypoints and QT; that is, if the watermarked image is displayed on the screen and captured by a camera, the watermark can be still extracted from the photo. A screen-shooting resilient watermarking algorithm should meet the following two basic requirements: robust keypoints and a robust watermark algorithm. In our case, we embedded watermarks by combining the feature region filtering model to SuperPoint (FRFS) neural networks, quaternion discrete Fourier transform (QDFT), and tensor decomposition (TD). First we applied FRFS to locate the embedding feature regions which are decided by the keypoints that survive screen-shooting. Second, we structured watermark embedding regions centered at keypoints. Third, the watermarks were embedded by the QDFT and TD (QT) algorithm, which is robust for capturing process attacks. In a partial shooting scenario, the watermark is repeatedly embedded into different regions in an image to enhance robustness. Finally, we extracted the watermarks from at least one region at the extraction stage. The experimental results showed that the proposed scheme is very robust for camera shooting (including partial shooting) different shooting scenarios, and special attacks. Moreover, the efficient mechanism of screen-shooting resilient watermarking could have propietary protection and leak tracing applications.

스마트폰 촬영 영상을 이용한 딥러닝 기반 위조지폐 판별

위조지폐 판별은 고가의 전용 장비를 사용하거나 고품질 스캐너를 사용한다. 하지만, 현재 스마트폰 카메라의 품질도 우수하여 충분한 해상도의 영상 획득이 가능하다. 본 논문에서는 스마트폰으로 촬영한 영상에 대해 딥러닝을 통해 위조지폐 진위를 판별하는 방법을 제안한다. 제안하는 방법에서는 합성곱 3계층과 전연결 2계층으로 구성된 합성곱 신경망 모델을 설계하여 최적화하였고, 학습데이터 수집 과정과 고화질 영상을 제한된 메모리에서 훈련하기 위한 패치 단위 처리 기술도 적용하였다. 또한, 패치 단위 판별 결과에 대해 최대 빈도수 기반으로 영상 단위로 위변조 판별하는 방법을 도입하였다. 실제 4개 프린터로 위조한 지폐를 이용한 실험에서 제안한 방법은 패치 단위 95.85%, 영상 단위 100% 정확도를 보였고, 이를 통하여 기존 UV나 광학 장비를 이용하지 않고도 스마트폰 촬영 영상만으로도 충분히 위조지폐를 판별할 수 있음을 보였다.

İnsansız Hava Araçları İçin Görüntü İşleme Tabanlı Otonom İniş

Günümüzde teknolojik anlamda herkes güvenli ölçümlerin hayatımızdaki etkisine oldukça aşinadır. Yani bu bağlamda araştırmacılar tarafından güvenli ölçüm çalışmaları için birçok çalışma yapıldı ve bunlardan biri İnsansız Hava Aracı. İnsansız Hava Aracının iyi bilinen bir kullanımı, belki de, bu cihazı sadece insansız hareketi için değil, aynı zamanda keyfi alanlar üzerinde uçuş sırasında benzersiz manevra için son derece pratik kılan güvenlik ve bakım ölçümlerindeki kabiliyetidir. Bu araştırmada İnsansız Hava Aracının otomatik inişi ele alınmıştır. İnsansız Hava Araçlarının (İHA) günümüzde gözetleme, keşif, askeri amaçlar, taşımacılık, zirai ilaçlama, kamera çekimi, yangın söndürme gibi birçok alanda kullanımı hızla artmıştır. İHA sistemlerinde kalkış, seyir, güdüm ve iniş kontrolü en önemli olaylardır. İHA’nın seyir halinde havadaki dengesi, çevre güvenliği ve uçuş stabilitesi açısından önem arz etmektedir. İHA’nın kalkış ve iniş manevraları güvenlik ve kararlılık açısından oldukça önemlidir. Bu çalışmada, havadaki bir İHA’nın görüntü işleme metodları kullanılarak zemin üzerindeki renkli bir alana otonom olarak inmesi sağlanmıştır. Alan üzerine dairesel biçimde bir cisim konulmuştur. Bunun için renkli dairesel cismin kamera ile koordinatları ölçülerek bu koordinatlara göre iniş yörüngesi belirlenmiştir. Kamera ile çekilen görüntü 512 piksele ayrılmıştır. Çekilen görüntünün tam orta noktası 256x256 pikseller olarak ayarlanmıştır. Kameradan gelen anlık geri kazançlar ile İHA kendi konumunu sürekli kontrol ederek sabit kalmaya çalışmıştır. Bu sayede otonom iniş için gerekli olan anlık kontrol edilen değerler kamera görüntüsü kullanılarak değerlendirilmiştir. Ayrıca çalışmada İHA için eşdeğer lineer bir transfer fonksiyonu elde edilmiştir. İHA’nın giriş sinyallerinin ve çıkış sinyallerinin değerleri incelenerek System identification ile uygun bir transfer fonksiyonu elde edilmiştir. Bu transfer fonksiyonu denetleyici tasarımında kullanılmıştır.

An exploding foil overpressure actuator for multipoint synchronous initiation and mach reflection generation

In this paper, two new forms (series-parallel and serial) of exploding foil overpressure actuators (EFOA) have been designed for the Mach reflection pressure studies and initiation of insensitive explosives at a small laboratory scale. These actuators, prepared with microelectromechanical system (MEMS) technique, can be in-situ integrated and mass-produced to reduce manufacturing cost. Electrical characterizations were carried out to obtained some important parameters, such as peak current and voltage, burst time and deposited energy, and the results revealed that the bridge foil of serial-type actuator has an earlier burst time and lower deposited energy than that of series-parallel one. Using an in-house built photon Doppler velocimetry (PDV), each flyer velocity history was measured separately, and the results indicated that the flyer speed of series-parallel actuator is higher, and its exit velocity consistency and time synchronicity are better than that of the serial ones. The flyers of two actuators exhibited a high level of integrity through the ultra-high-speed camera shooting. Based on the Lagrange hydrodynamic equation group, one-dimensional simulation was conducted by introducing the equation of state, discharging circuit equation and Joule heat equation to numerically simulate the whole process of actuator action. The simulation results are well agreed with the experimental ones both in electrical parameters of bridge foil and flyer velocity. Finally, the multipoint synchronous initiation of HNS-IV explosives was successfully achieved with the actuator. All the efforts demonstrated that the actuators are feasible and attractive.

Deep Template-Based Watermarking

Traditional watermarking algorithms have been extensively studied. As an important type of watermarking schemes, template-based approaches maintain a very high embedding rate. In such scheme, the message is often represented by some dedicatedly designed templates, and then the message embedding process is carried out by additive operation with the templates and the host image. To resist potential distortions, these templates often need to contain some special statistical features so that they can be successfully recovered at the extracting side. But in existing methods, most of these features are handcrafted and too simple, thus making them not robust enough to resist serious distortions unless very strong and obvious templates are used. Inspired by the powerful feature learning capacity of deep neural network, we propose the first deep template-based watermarking algorithm in this paper. Specifically, at the embedding side, we first design two new templates for message embedding and locating, which is achieved by leveraging the special properties of human visual system, i.e., insensitivity to specific chrominance components, the proximity principle and the oblique effect. At the extracting side, we propose a novel two-stage deep neural network, which consists of an auxiliary enhancing sub-network and a classification sub-network. Thanks to the power of deep neural networks, our method achieves both digital editing resilience and camera shooting resilience based on typical application scenarios. Through extensive experiments, we demonstrate that the proposed method can achieve much better robustness than existing methods while guaranteeing the original visual quality.