288results about How to "Flexible detection" patented technology

A feature extracting unit obtains sensor data from a plurality of sensors to calculate each feature. When an event determining unit determines the occurrence of an event based on each feature, a display data constructor generates remote-controller display data for displaying the event, and controls a remote-controller display device to display the remote-controller display data. When a user decision is input from a user input IF based on this display, a control unit controls the sensors to be turned ON or OFF. When an infrared sensor detects an abnormality, a microwave sensor whose power consumption is small after the infrared sensor is turned ON. When the microwave sensor detects an abnormality, a video camera and a microphone are turned ON, and the microwave sensor is turned OFF. A communication unit wirelessly transmits an image signal captured by the video camera and an audio signal processed by the microphone. Then, if the infrared sensor does not detect an abnormality, the video camera and the microphone are turned OFF. With this arrangement, power consumption can be suppressed. The present invention is applied to, for example, a security system, for example, for monitoring outside a vehicle by a video camera disposed in the vehicle when the vehicle is parked.

The invention discloses a network security defense system based on a software-defined network and a working method of the network security defense system. The network security defense system comprises an SDN controller, an IDS decision making server and an IDS device. The IDS device is suitable for spot-checking messages, and in other words, when the IDS device detects the messages with the DDoS attack characteristics, the messages are reported to the IDS decision making server; the IDS decision making server formulates a processing strategy corresponding to the messages with the DDoS attack characteristics according to reported information, and the processing strategy is issued to the SDN controller for threat processing. The messages are detected in a spot-checking mode, the burden on the SDN controller is greatly reduced, the burden on the server is further reduced in the mode that detection and decision making are separated, and the system is more suitable for network transmission of high-flow data.

The invention discloses a safety enhanced virtual machine communication method and a virtual machine system. The safety enhanced virtual machine system according to one embodiment comprises a plurality of virtual machines positioned on a same physical machine in a virtual network. The virtual machine system comprises a virtual network safety manager coupled in the virtual network and a shadow system coupled with the virtual network safety manager, wherein the shadow system is positioned on the other physical machine different from the physical machine in which the virtual machines are positioned, the virtual network safety manager copied packets are sent among the virtual machines and the copied packets are forwarded to the shadow system through a physical safety device; the shadow systemreturns response information indicating the received packets to the virtual network safety manager; and the virtual network safety manager sends the packets among the virtual machines according to the returned response information.

The invention relates to a crop pest detection system positioned on an unmanned aerial vehicle. The crop pest detection system comprises a random access memory, a high-definition camera, an image processor and a main controller, wherein the high-definition camera is used for capturing a crop image; the image processor is connected with the high-definition camera and is used for processing the crop image to obtain a crop pest type in the crop image; the main controller is connected with the random access memory and the image processor respectively and is used for storing the crop pest type in the random access memory. By virtue of the crop pest detection system positioned on the unmanned aerial vehicle, the pest type of a field crop can be automatically obtained, and relevant agricultural management departments can conveniently call the system when the unmanned aerial vehicle returns, so that targeted pest prevention and control measures are taken.

The invention relates to an automatic filling device and a control method of the automatic filling device. The automatic filling device comprises a detergent storer, a filling amount detecting container, a controller, and a switch valve module, wherein the filling amount detecting container is used for temporarily storing the filled detergent, detecting and feeding back the filling amount of the detergent in a real time manner; the controller is used for receiving the feedback signal from the filling amount detecting container and judging whether the filling amount of the detergent is up to the preset filling amount; the switch valve module is controlled by the controller to allow the detergent storer to be communicated with or separated from the filling amount detecting container; when the detergent filling amount is not up to the preset filling amount, the detergent storer is communicated with the filling amount detecting container, and the detergent flows from the detergent storer to the filling amount detecting container; and when the detergent filling amount is up to the preset filling amount, the detergent storer is separated from the filling amount detecting container. The automatic filling device can accurately control the detergent filling amount, be suitable for the fields of household appliances and industrial application, especially for washing machines or dish-washing machines.

A feature extracting unit obtains sensor data from a plurality of sensors to calculate each feature. When an event determining unit determines the occurrence of an event based on each feature, a display data constructor generates remote-controller display data for displaying the event, and controls a remote-controller display device to display the remote-controller display data. When a user decision is input from a user input if based on this display, a control unit controls the sensors to be turned ON or OFF. When an infrared sensor detects an abnormality, a microwave sensor whose power consumption is small after the infrared sensor is turned ON. When the microwave sensor detects an abnormality, a video camera and a microphone are turned ON, and the microwave sensor is turned OFF. A communication unit wirelessly transmits an image signal captured by the video camera and an audio signal processed by the microphone. Then, if the infrared sensor does not detect an abnormality, the video camera and the microphone are turned OFF. With this arrangement, power consumption can be suppressed. The present invention is applied to, for example, a security system, for example, for monitoring outside a vehicle by a video camera disposed in the vehicle when the vehicle is parked.

The invention provides a synchronous and indirect competitive immunological detection method and a kit of plural small molecular compounds. The immunological detection method is a liquid phase immunological detection method which uses an encoded microsphere as a solid phase carrier and phycoerythrin as a fluorescent marker and is used for competitive specificity reactions of a small molecular compound antigen. Firstly, a captured antigen is covalently bound to the surface of the encoded microsphere, an indirect competitive fluorescent immune complex of microsphere-captured antigen-detected antibody-second antibody-phycoerythrin is formed in a filter film plate through capturing the antigen, detecting the antibody and binding the second antibody specificity, and then the indirect competitive fluorescent immune complex flows across a suspension chip or a detection region of a flow analysis system one by one, recognizes different encoded microspheres and detects the fluorescent intensity of the phycoerythrin to complete the detection. Different captured antigens are connected by using the different encoded microspheres, a plurality of small molecule antigens in the same sample can be synchronously detected, and one or more small molecule antigens to be detected in different samples can also be detected. The invention has the advantages of fast speed and high flux.

The invention discloses a channel detection indication method, a terminal and network equipment. The method comprises the following steps: detecting a first physical downlink control channel PDCCH; and if the scheduling downlink control information DCI is received from the first PDCCH, determining whether to detect the second PDCCH or not according to the scheduling DCI. According to the embodiment of the invention, whether the PDCCH is detected or not is dynamically indicated by scheduling the DCI, so that the PDCCH detection of the terminal is more flexible, unnecessary PDCCH detection can be avoided, and the power saving of the terminal is facilitated.

The invention discloses a detection method suitable for the call receiving and making behavior of a driver under multiple postures. The detection method comprises the following steps: 1; image acquisition: acquiring the face image information of the driver; 2. image recognition: carrying out real-time face detection on the acquired image; 3. face feature point calibration: carrying out feature point positioning on the detected face, wherein the feature points comprise eyebrows, eyes, a nose, a mouth and the like; 4. the three-dimensional posture detection of the head: utilizing geometrical features to detect the three-dimensional posture of the head, and judging the posture of the head; 5. calibration of a face outline ellipse: utilizing the sign feature points including the eyebrows, the eyes, the nose and the mouth and the three-dimensional posture of the head of the driver to carry out the calibration of the face outline ellipse on the driver through linear regression; and 6. carrying out call receiving and making behavior detection and judgment out of the calibrated face outline ellipse. Through the three-dimensional posture detection of the head of the driver, a supervised gradient descent method is adopted to position face feature points, and the calibration of the face outline ellipse is combined to carry out judgment so as to realize the flexible detection, the intelligent analysis and the accurate judgment of the call receiving and making behavior of the driver under multiple postures at multiple angles, and the detection method is high in instantaneity, practicality and reliability.

The invention discloses a control method, a control device and a system used for a vehicle-mounted radar. The vehicle-mounted radar is rotationally installed on a vehicle through a drive device. The control method includes the steps that a signal used for presenting the rotation angle of a steering wheel is received; according to the rotation angle of the steering wheel, a target angle of the vehicle-mounted radar is determined; according to the target angle of the vehicle-mounted radar, the drive device is controlled to work, and accordingly the vehicle-mounted radar rotates to the target angle. By means of the control method, a detection area of the vehicle-mounted radar is controlled according to the current rotation direction of the vehicle, so that the detection area of the vehicle-mounted radar adapts to the rotation direction of the vehicle. Thus, when the vehicle runs at a bend, the vehicle-mounted radar can flexibly and accurately detect the distance between the vehicle and a front vehicle or the distance between the vehicle and a rear vehicle or the distance between the vehicle and surrounding obstacles, and accordingly accidents are reduced.

The invention discloses a detection device for knife point dynamic characteristics of a numerical control machine tool and a method of the detection device. Mainly the problems that detection performance indexes of knife point dynamic characteristics of an existing numerical control machine tool are not comprehensive or accurate, and detection cost of instruments and cutting test pieces is high are solved. The detection device for the knife point dynamic characteristics of the numerical control machine tool comprises a supporting base, a connecting plate and an L-shaped supporting wall plate. The top of the supporting base is provided with displacement sensors in a penetrating manner through holes, wherein the displacement sensors are arranged in the manner that every two displacement sensors are perpendicular to each other; meanwhile, acquired temperature data of main heat sources of the machine tool are integrated so that a main shaft can acquire real-time location and posture information of a point of a knife point in the swinging or rotating process, and correlation calculation of accuracy and generation of compensating data are completed by automatically obtaining fused data information through a data collector and transmitting the fused data information to a computer at high speed to be recorded and analyzed. The detection device is ingenious in structural design, the knife point characteristic testing functions are comprehensive, universality is high, numerical control machining precision testing cost is greatly reduced, and good practical and popularization value is achieved.

The invention provides an integrated pulse eddy current sensing infrared thermal imaging detection apparatus and a detection method thereof. The integrated pulse eddy current sensing infrared thermal imaging detection apparatus comprises an electromagnetic induction actuator for generating an alternating magnetic field, an infrared thermal imager and an embedded industrial control plate, wherein one end of the actuator coil of the electromagnetic induction actuator is connected to the embedded industrial control plate, the other end is aligned with the crack defect on a specimen, the infrared thermal imager is connected to the embedded industrial control plate, and the infrared thermal imager collects the infrared image of the specimen and transmits the image to the embedded industrial control plate. According to the present invention, the pulse eddy current sensor is used as the excitation source, and the excitation of the infrared thermal imaging detection and the collection device are integrated, such that the structure design is compact and reasonable, the single-hand operation can be achieved, and the field practicability is strong; the synchronization of the infrared image acquisition and the actuator excitation is achieved by the industrial control tablet PC, such that the operation of the apparatus is simple and convenient; and the infrared detection method provided by the embodiments can achieve the simple and rapid detection of various defects.

The invention discloses a firewall based on intrusion detection system feedback in a cloud environment and an implementation method thereof. The firewall comprises a firewall part and an IDS part. Thefirewall part filters the network data for the first time according to a filtering rule; the IDS part further filters the data passing through the firewall and screens the data based on an IDS filtering rule; and a filtering strategy of the firewall part is updated through feedback information of the IDS part in the system. A firewall and an intrusion detection system (IDS) are combined, interlocking is implemented through the IDS and the firewall, attack defense is divided into two parts, and defense work is naturally divided into two parts. The firewall part is used for coarse-grained defense, the IDS part is used for fine-grained analysis of grouping information, double-layer defense is more effective and accurate, defense attacks are separated from a cloud service server, the qualityof cloud services is not affected, and the response delay of the cloud services is reduced; and an IDS-based feedback change rule is adopted, so that attacks can be flexibly detected.

The invention relates to a spherical underwater robot used for outdoor water quality monitoring, characterized in that the spherical underwater robot includes: a mechanical seal, a shaft, a stud, a motor waterproof cover, front and rear motor splints, a spherical shell, and an end cover , fairing, splint, snap ring, motor seat, front and rear of the motor seat and circuit board; the circuit board is placed in a spherical shell; the motor unit is composed of a motor, a motor box, a motor splint and a motor seat; the mechanical seal is fixedly connected The motor is on the shaft; the shaft is connected to the blade; the motor is placed in the motor box; the motor box is composed of an end cover, a motor waterproof cover and a fairing; the motor is fixed to the end cover by studs; The waterproof cover of the motor is fixed on the snap ring through the front and rear motor splints and the motor seat; the snap ring is fixed on the largest part of the circumference of the spherical shell. The spherical underwater robot provided by the embodiment of the present invention has flexible movement, can turn freely, and has small movement resistance. And the cost is low, and it is suitable for waters of various depths.

The invented polynary immunoassay plate relates to an enzyme immunoassay chip reaction device which is applied widely to the medical detection, analysis and other field. The polynary immuoassay plate has several holes with antigen fixed by chemical bond to an orifice plate, the surface of the orifice plate as solid carrier is closed and the sample to be tested is added to produce hybridized reaction with the biomolecules fixed to the orifice plate. After unreacted matter is washed off, the labeled antigen is added for hybridization, and the unreacted matter is washed off again before the label signal is detected. Several indexes may be detected simultaneously. The present invention is flexible and saving in tested sample amount, and some commercial autoamtic instruments may be used.

The invention discloses an android malicious code detection system and method based on Opcode backtracking. The method includes the steps that firstly, a rule base is established, and detection strategies are customized by the rule base according to an object to be detected; an API called initially is placed in a queue as a name to be matched; the name to be matched is read from the queue, and names of functions called by all Opcode instructions are obtained by using a dex analysis and Opcode disassembling module; the names of the called functions are contrasted with the name to be matched, the name of the function where the successfully matched Opcode instruction belongs is placed in the queue as a name to be matched, the operations are repeated until the queue is empty, whether a called relational tree conforms to detection rules customized in the rule base or not is analyzed and judged on the basis of the rule base, and if yes, the relational tree is a malicious sample; otherwise the relational tree is safe. According to the Android malicious mode detection system and method based on Opcode backtracking, the problems of false alarm and alarm failure in an existing malicious code detection method can be well solved.

The invention belongs to the field of wall-climbing robots, in particular to a foot-type magnetic adsorption wall-climbing robot, which is composed of a carrying platform, a walking leg and an adsorption foot. Various actuators can be mounted on the carrier platform to perform various operations on the iron-based wall surface. Each walking leg comprises a two-degree-of-freedom series mechanism, aservo-electric cylinder and a ball hinge, wherein the end of each walking leg is connected and absorbed sufficiently. The suction foot adopts a screw nut mechanism to adjust the height of the permanent magnet and the wall surface so that the suction force can be adjusted. Robots can be thought of as combinations of 4-RRR Planar mechanisms and elevating mehcanisms. the movement of the two mechanisms is controlled to realize the omni-directional movement of the platform in the space. Robot can move straight, turn and adjust posture flexibly on the different curvature of the wall. The invention solves the shortcomings of unreliable adsorption and inflexible steering of the conventional magnetic adsorption wall climbing robot, and has the advantages of simple structure, strong load capacity, good obstacle-crossing ability, flexible movement and reliable adsorption.

The invention provides a submerged type pipeline welding joint polishing robot based on a modular design. The submerged type pipeline welding joint polishing robot comprises a power module and a polishing module. The power module is used for driving the whole robot to move in a pipeline. The power module is divided into an upper portion, a middle portion and a lower portion to adapt to the change of a pipe diameter. The polishing module is used for polishing the welding joint position of the pipeline. The polishing module is composed of three parts which are a front and back supporting structure, a main shaft and a polishing feeding operating mechanism. The power module and the polishing module are connected through a cardan joint to guarantee the coordination of the two modules when the robot moves forward and backward and guarantee that the robot can move normally when passing through the curved pipeline. According to the submerged type pipeline welding joint polishing robot, the modular design is adopted, and modules such as a detecting module and a spraying module can be developed flexibly in the later period.

A slow attack detection method and apparatus are disclosed. The slow attack detection method is characterized in that the method comprises: determining a preset attack characteristic for calculating an attack value according to a preset attack value calculation rule; Determining values of each preset attack characteristic in the received message, and calculating an attack value of the message according to the characteristic value; Comparing whether the calculated attack value is greater than the preset attack threshold; Determining the the message is a slow attack message when the calculated attack value is greater than a preset attack threshold value, and calculating a new attack threshold value; wherein the new attack threshold is used for subsequent slow attack detection, and the new attack threshold is not greater than the old attack threshold.

The invention relates to a traffic information management system and method, and the system comprises: information identification bar codes, for storing vehicle and / or driver ID information; information identification and receive-transmit device, for identifying the ID information and transmitting the information to a management center server and obtaining vehicle and / or driver related information from the server; wireless network system, for implementing the interaction between the information identification and receive-transmit device and the management center database; management center server, comprising database, processing unit and network unit for storing driver and / or vehicle IDs information, managing the information and reading and modifying the database, as well as transmitting the information to the information identification and receive-transmit device through the network unit. The invention can load information bar codes on the existing driver and vehicle ID papers and traffic information management and field penalization by handsets with information identification units through ordinary wireless communication networks.

The present invention discloses a slope detection method and a slope detector. The slope detector comprises an atmosphere sensor and a control part. The control part confirms whether the driving rod of the vehicle is a slope according to the atmosphere data input from the atmosphere sensor. The control part comprises the following components: a height calculation part which calculates the height according to the atmosphere data obtained from the atmosphere sensor each time the vehicle is driven for a preset unit driving distance; a slope gradient calculation part which calculates the slope gradient of the road according to the difference between the height calculated last time and the height calculated this time and the unit driving distance; a counting part which counts the time of slopegradient calculation; and a slope confirming part which confirms that the road is a slope when the slope gradients obtained by the slope gradient calculation part are continuously in a preset range in the preset number of times.

The invention relates to an apparatus for detecting resistance welding of a battery pack and a method thereof. The apparatus includes: a base seat, and a battery pack positioning frame and a fixing support, wherein the battery pack positioning frame and the fixing support are arranged on the base seat. The fixing support is provided with two translation mechanisms with each provided with a lifting mechanism. The lifting mechanism is provided with a detection probe which faces vertically downward. The battery pack positioning frame includes an L-shaped frame and a baffle plate which is slidably connected to a backside plate of the L-shaped frame. The baffle plate and a passing driving mechanism are connected. The translation mechanisms, the lifting mechanisms and the passing driving mechanism are separately in electrical connection with a control system. The detection method includes: the translation mechanisms adjusting horizontal positions of the detection probe, the passing driving mechanism adjusting the size of the battery pack positioning frame, and the lifting mechanisms controlling the lifting of the detection probe, the control system automatically determining whether the battery pack qualifies and sending alarm based on the determination. The apparatus and the method of the invention are accurate in positioning, flexible and reliable in detection, time and labor saving, excellent versatility, high security and work efficiency.

The invention discloses a color image edge detection method which comprises the following specific steps: extracting each color channel image, generating color antagonism images, calculating an edge information distribution image, and refining the color image edge. According to the method, a color image is discomposed into a red channel image, a green channel image, a blue channel image and a yellow channel image, red-green antagonism images and blue-yellow antagonism images are calculated respectively, edge information distribution images are obtained on the antagonism images through calculating, and the strength of the color edge and the brightness edge are adjusted conveniently through two groups of antagonism weighting factors introduced in the process of adjusting, calculating and generating the color antagonism images, thus the color image edge can be detected flexibly and effectively. The color image edge detection method has the advantages that the brightness or color edge information of the nature color image can be effectively detected through simple parameter selection, and in addition, the brightness and color edge information of the nature color image can be selectively extracted.

The invention discloses a wireless video monitoring system based on a wide water domain. A ship, vehicle and boat hybrid monitoring manner is adopted to establish a monitoring system, a wireless-based network-carrying system is constructed for a ship-vehicle-boat monitoring system, wireless network coverage is carried out by a 3G (The 3rd Generation Telecommunication)+WiMAX (Worldwide Interoperability for Microwave Access)+WiFi (Wireless Fidelity) hybrid architecture mode, servers are erected on a boat, a vehicle and a boat for transmitting and processing videos, and video data acquired by the ship, the vehicle and the boat are transmitted to a monitoring center through the network-carrying system. The monitoring system based on the wide water domain, provided by the invention, has the advantages of good coverage type, flexibility in detection, no influence from weather environment, and the like, can better avoid the defect brought by traditional satellites, can be used for monitoring the wide water domain in real time, has a aerial panorama with the formation time reduced from traditional 3-7 days to about 1-2 days, and can work under weather conditions such as rain, snow, fog and the like that the traditional satellites can not form an image and give a corresponding result image.

The present invention discloses a life detecting system and method based on a multi-rotor UAV. The system comprises a multi-rotor UAV and a life detecting radar device mounted on the multi-rotor UAV,and further comprises a motion compensating device connected with the life detecting radar device. When the multi-rotor drone is flying in a to-be-tested area, the life detecting radar device performsliving body target detection; and during the detection process of the life detecting radar device, the motion compensating device performs motion compensation on a radar detection signal according tothe real-time position information of the multi-rotor drone so as to compensate for the displacement deviation caused by the jitter of the multi-rotor UAV. The technical scheme of the present invention has the advantages of a simple structure, a low cost, dynamic life detection, high detection efficiency, high precision, a flexible detection mode, and the like.

The invention discloses a construction method for a conditional overexpression Yap1 gene mice model at an H11 locus. The construction method comprises the steps: firstly, obtaining an inducible Yap1 overexpression transgenic vector; then, obtaining transgenic F0-generation mice; next, enabling the F0-generation mice to mate with wild-type mice, carrying out passage and line establishment, and thusobtaining F1-generation mice; and carrying out expanding propagation of positive F1-generation mice, then mutually mating, and thus obtaining the conditional overexpression mice Yap1 animal model. The transgenic mice model constructed by the method can be used for studying the function of Yap1 in different tissues and different developmental stages and the action mechanism of Hippo signaling pathways, can also be used for studying the function of the Hippo signaling pathway in animal development, biotumorigenesis, regenerative medicine and other disciplines, and has important guiding significance on biological signal transduction studies.

The invention disclose a simulated obstacle crossing probe vehicle, which consists of an obstacle crossing system and a probe system, wherein a first bracket and a second bracket are fixed on a box body of the obstacle crossing system; a storage battery and a circuit board are arranged inside both sides of the box body respectively; connecting rods are connected to the brackets through second connecting rod shaft sleeves and pin shafts; one ends of connecting swing stems are movably connected to the connecting rods, and the other ends of the connecting swing stems are supported on middle tires; motors are connected with wheels through motor shafts, snap springs and first connecting rod sleeve shafts; the front part of the box body is connected to the a steering motor through first connecting rods, second connecting rods and third connecting rods; and the motors are connected tire frames through the motor shafts, bearings, the snap springs and gaskets. A probe head is fixed at the front end of a direct-current stepping motor of the probe system; the direct-current steeping motor is movably connected with a straight-teeth cone gear; a corner connecting rod is connected with the direct-current stepping motor and a shaft sleeve; a motor connecting rod is connected with another shaft sleeve through a screw; and a plurality of the structures are combined and connected to form a trunk-shaped structure which is fixed on the box body through a first cone gear shaft, a second cone gear shaft and a corner base.

The embodiment of the invention provides a wafer detection method, device and equipment. The method comprises the following steps: acquiring an image of a to-be-detected wafer; comparing each to-be-detected area with a corresponding area in the standard wafer, wherein the standard wafer comprises N areas, each area of the standard wafer is in one-to-one correspondence with each to-be-detected area, and the position of the to-be-detected area in the to-be-detected wafer is the same as the position of the corresponding area in the standard wafer, and if the difference between each to-be-detectedarea and the corresponding area of the standard wafer is within a preset range, determining that the to-be-detected wafer is a qualified wafer. The judgment standards of different areas in the waferfor defects are different, and defect detection is performed on each to-be-detected area so that the defect detection of the wafer is more flexible, and the detection accuracy is higher. The problem in the prior art that when defects of the wafer are detected, the detection standard for detecting whether the defects exist in the wafer is inflexible and the detection result of whether the defects exist in the wafer is inaccurate is solved, and thus the production cost is reduced.

The invention provides a digital flat panel ray detection ray source end robot based on Mecanum wheels. The digital flat panel ray detection ray source end robot based on the Mecanum wheels comprises a frame I, Mecanum wheels I, servo motors I, a front tracking sensor I, a rear tracking sensor I, an X-ray source, permanent magnets I and a motion control box I, the motion control box I is arranged on the frame I, the front tracing sensor I is arranged at the front end of the frame I, the rear tracking sensor I is arranged at the rear end of the frame I, the X-ray source is arranged in the middle of the frame I, the Mecanum wheels I are arranged on the two sides of the frame I respectively, the servo motors I are connected with the motion control box I, and the permanent magnets I are arranged on the two sides of the bottom of the frame I respectively. The detection robot can conduct omni-directional movement, and on one hand, advancing and retreating, lateral moving and turning around own center can be achieved; on the other hand, rotational movement around any point can be further achieved; the flexibility of the robot on weld joint detection on the curved surface of equipment can be greatly improved specially in large-scale in-service bearing special equipment such as power station boilers, spherical tanks and vertical storage tanks.

Provided is a high-resolution remote-sensing image ship target detection method based on deep learning. The method includes steps: acquiring a high-resolution remote-sensing image including a ship target, and performing artificial marking on the ship in the image by employing a minimum quadrangle capable of surrounding the ship target; setting a structure of a deep learning network model, connecting feature maps with different semantic properties and resolutions, then setting multiple reference quadrangles for positions of the feature maps, and classifying the reference quadrangles by employing a full connection layer; and training the network model, inputting a to-be-detected high-resolution remote-sensing image to the trained network model, detecting a ship area in the image, and constructing a minimum quadrangle surrounding the ship target so that the accurate position of the ship target is indicated. According to the method, compared with the conventional ship target detection method, the interference of environment factors can be more effectively resisted, and ships in different locations and with different dimensions and attitudes in the high-resolution remote-sensing image can be detected more stably and accurately.