183 results about "Camera identification" patented technology

The invention discloses a picking machine and a control method. The picking machine comprises a robot body, a fruit diameter separating device and a mechanical arm, wherein the fruit diameter separating device and the mechanical arm are installed on the robot body. The laser radar and the 2D camera are integrated on the robot body, and the depth camera is integrated on the mechanical arm. The processor is in communication connection with the laser radar, the 2D camera and the depth camera; the laser radar scans environment point cloud, the processor constructs a grid map and plans a driving path, mature fruits on the corresponding side are recognized through the depth camera in the driving process, the poses of the mature fruits in a mechanical arm coordinate system are determined, and the mechanical arm is controlled to pick the mature fruits. And the collected fruits are placed in an inlet area of the fruit diameter separation device. The picking robot integrates the functions of environment perception, autonomous navigation, automatic identification and picking of mature fruits and simultaneous automatic classification of fruits with different diameters, so that the picking efficiency is improved while the fruit picking process is greatly simplified.

The invention relates to the technical field of automatic production, in particular to an automatic unstacking method for air conditioner condenser cores. According to the technical scheme, the method comprises the following steps that a 3D camera is arranged above a condenser core feeding platform, and the side faces of cooling fins of a condenser core are recognized; a robot is used for driving a clamp to clamp and move the uppermost condenser core to a discharging conveying line according to the two fin side faces, recognized by the 3D camera, of the condenser core; a 2D camera is arranged on the discharging conveying line to recognize and detect the end type of the condenser core, and finally the robot obliquely places the condenser core on the discharging conveying line in the direction that the U-shaped pipe end faces downwards according to the end type of the condenser core. The method has the beneficial effects that the recognition efficiency can be improved on the premise of ensuring the recognition accuracy; and in addition, a 3D recognition method and a 2D recognition method are optimized, the recognition efficiency is further improved, and the automatic unstacking effect is improved.

The invention discloses a nozzle voltage adjustment method and device based on camera automatic identification printing color blocks and a storage medium, and relates to the technical field of printing device calibration, and the method comprises the steps: controlling a printing device to output a pattern containing each nozzle independent test area and color blocks; acquiring chromaticity distribution data through an image acquisition device, and extracting characteristic parameters such as brightness and saturation; determining a reference characteristic parameter based on a reference selection rule, and calculating a deviation value between each test area and the reference; an adjustment instruction is generated according to the mapping relation between the deviation value and the voltage adjustment amount, verification printing and iterative optimization are carried out after implementation until the precision is met, and through dynamic selection of a reference color block, LAB color space chromatic aberration calculation, voltage sensitivity coefficient matrix mapping and aging compensation, the adjustment precision and efficiency are improved, the dependence on artificial experience is reduced, and the adjustment precision and efficiency are improved. The method is suitable for nozzle calibration of high-precision printing equipment.

The invention discloses a wire harness laser coding system based on CCD camera recognition. The wire harness laser coding system comprises a mechanical arm, a wire harness clamping jaw, a wire harness end clamping unit, a horizontal guide rail, a ccd visual detection unit and a laser coding device. The sides, close to each other, of the two wire harness end clamping units are coaxially provided with two wire harness end insertion ports. The two wire harness clamping jaws are arranged on the displacement mechanical arm, the two wire harness clamping jaws can clamp the positions, close to the two ends, of the target wire harness respectively, and drive the clamped target wire harness to move to the position between the two wire harness end clamping units under the action of the mechanical arm, so that the two ends of the target wire harness are coaxially aligned with wire harness end inserting openings of the two wire harness end clamping units respectively; the two wire harness end clamping units can synchronously move in the extending direction of the horizontal guide rail. Original characters and patterns on the target wire harness are made to deviate from the sight line of the ccd visual detection unit through the ccd visual detection unit, and therefore finally achieved code printing effectively avoids the original characters and patterns on the surface of the wire harness.

The invention relates to a video quality detection method and system, an electronic device and a storage medium, and is used for a reverse proxy server, the reverse proxy server is deployed at a cloud, and the method comprises the following steps: obtaining camera identification information reported by a proxy client; when a real-time streaming transmission protocol request containing a virtual real-time streaming transmission protocol port initiated by the PC client is received, forwarding the request to a corresponding proxy client according to camera identification information corresponding to the virtual real-time streaming transmission protocol port; the virtual real-time stream transmission protocol port establishes a mapping relationship with the camera identification information in advance; receiving video data uploaded by the proxy client; the video data is locally acquired by the proxy client from the camera according to the forwarding request; and encapsulating the video data into a standard real-time stream transmission protocol response, and returning the real-time stream transmission protocol response to the PC client for video quality detection. A communication link can be established for video transmission, the process is simplified, and the video quality detection efficiency is improved.

The application relates to the field of agricultural automation equipment, in particular to an apple picking robot based on visual guidance and a picking method, which recognizes apples and determines position information by using a camera, controls a mobile chassis to move in a large range according to position information of a control system, realizes height and angle adjustment in combination with a lifting column and a small arm, and then picks apples by using a fruit picking opening and conveys the apples to a collecting frame by using a mesh belt. The application greatly reduces the labor intensity of manual picking and improves the picking efficiency. Then, real-time pose information of the apples is acquired by using a camera, a mechanical arm is guided to adjust actions to complete apple tracking and grabbing, so that accurate and efficient picking work is realized, the labor intensity of manual picking is reduced, and the picking efficiency is improved. The robot can automatically recognize target apples and pick the apples, so that the picking efficiency and productivity are improved.

This invention relates to the field of measurement data analysis technology, specifically to an engineering measurement data analysis method and system based on image recognition. The method includes the following steps: using an industrial camera to identify structural edges, measurement markers, and calibration rulers to form a reference set of measurement point coordinates; establishing corresponding quantities of directional scales based on the calibration ruler direction; combining the displacement relationship of the measurement points to form a measurement point interval characterization set; determining the edge deviation reference rate; and finally forming a boundary positioning association result. This invention uses grayscale abrupt changes to define edges and construct measurement point positioning references; calculates the lateral and longitudinal displacements of the calibration ruler to extract corresponding directional scales; combines measurement point displacement and edge direction to form interval characterizations; analyzes the normal span and surface deviation state to form edge references; and combines the boundary extension orientation to define the associated segments of the arrangement trajectory. This reduces interference from manual interpretation, maintains consistency of multi-directional conversion benchmarks, controls the error accumulation process, and improves the coherence and stability of engineering monitoring data processing.

The utility model relates to the technical field of mask production lines, and discloses a mask observation mechanism of a large-size mask production line, which comprises a blanking robot and a blanking clamping jaw arranged at the movable end of the blanking robot, and the blanking clamping jaw is provided with a camera identification assembly matched with the movable clamping jaw. The camera recognition assembly is used for recognizing and positioning the position of a Box, an observation table is arranged on one side of the discharging robot, a vertical plate clamping assembly is arranged on the observation table, and the vertical plate clamping assembly comprises a pair of vertical beams and a supporting base, the vertical beams can synchronously move face to face or back to back, and the supporting base can move in the vertical direction. When the mask placing device is in butt joint with an upstream discharging robot, the mask is in a vertical state, the dust falling risk is small, workers can observe the mask face and the mirror face on the two sides of the mask respectively, manual face turning is not needed, the mask placing precision is greatly improved, the collision risk is small, the burden of the workers is relieved, and the production efficiency is improved.

A method of recognizing a barcode by using a barcode reader (BCR) camera installed in a conveyor environment includes: obtaining, by using a BCR camera, a barcode image of a barcode on an object on a conveyor; obtaining a field of view of the BCR camera based on a conveyor environmental condition and a BCR camera setting condition; obtaining a moving distance of the object on the conveyor by using the field of view of the BCR camera, the conveyor environmental condition, and the BCR camera setting condition; generating a blur kernel of the barcode based on a moving direction of the object and the moving distance of the object; and deblurring the barcode image by using a pre-trained deblurring algorithm based on the blur kernel.

The invention discloses a multi-source fusion near-electricity construction area dynamic safety early warning method and system, and the method comprises the steps: dividing a construction area into different voltage levels based on an electric field sensor, and obtaining the positions of all electrified bodies; using a laser radar to detect and calculate the position and the speed of a reflector, comparing the position and the speed with the positions and the speeds of the personnel and the equipment identified by the camera, and matching the reflector with the identified personnel and the equipment; calculating the relative included angle, the change rate of the relative included angle, the distance, the change rate of the distance and the direction transformation entropy, and calculating the intrusion probability of personnel and equipment; and performing graded early warning according to the number of personnel, the number of equipment and the intrusion probability of the personnel and the equipment in the regions with different voltage grades. According to the method, the influence of the uncertainty of the moving trajectory on safety evaluation is considered, graded early warning is realized in combination with the regional voltage grade and the intrusion probability, and the early warning accuracy of construction safety management is improved.

The invention discloses a method and a device for separating gum-producing dandelion leaves and roots, and belongs to the technical field of agricultural automation equipment. The device comprises a rack, a leaf clamping conveying belt, a root clamping conveying belt, an adjustable cutting knife, a flexible tensioning device, a binocular vision camera recognition device and a scattered leaf collecting fan. According to the working process, the leaf clamping conveying belt clamps plant leaves and conveys the plant leaves in a directional mode; after the plant reaches a preset station, the root clamping conveying belt cooperates with the flexible tensioning device to clamp the root, and pre-tightening force is applied to the whole plant; the binocular vision camera recognition device recognizes the optimal cutting position of the blade root joint part in real time; the adjustable cutting knife executes a cutting action at a specified position according to an identification result to realize precise separation of a leaf part and a root part; the scattered blade collecting fan collects scattered blades through negative pressure. According to the device, efficient, accurate and automatic operation of separating the leaves and the roots of the gum-producing dandelions is achieved, and the problem that traditional manual separation is low in efficiency is effectively solved.

The utility model discloses a robot-based sucking and feeding device compatible with various pipe fittings, which comprises an overall fixing frame, a feeding belt conveying line, a chain plate conveying line, a carrier mechanism, a connecting belt conveying line, a backflow belt conveying line, a camera recognition mechanism and a robot sucking mechanism, the feeding belt conveying line, the chain plate conveying line, the connecting belt conveying line and the backflow belt conveying line are arranged on the overall fixing frame, the outlet end of the feeding belt conveying line is connected with the connecting belt conveying line, and the outlet end of the connecting belt conveying line is connected with the inlet end of the backflow belt conveying line. A carrier mechanism is arranged on the chain plate conveying line, and the camera recognition mechanism and the robot suction mechanism are arranged above the feeding belt conveying line. The pipe fitting conveying device can convey pipe fitting products, automatically recognize the pipe fitting products and finally suck and place the pipe fitting products to convey the pipe fitting products according to fixed postures, is compatible with various types of pipe fittings, and is good in carrier compatibility.

A method is provided of determining a blind spot of a display module caused by a steering wheel. The method includes a synchronization operation of allowing, by a control unit, one or more cameras installed on a display module to identify a driver’s eyes, a blind spot determining operation of driving, by the control unit, cameras among the one or more cameras, of which views are blocked by the steering wheel and which cannot identify the driver’s eyes and identifying the driver’s eyes, and a blind spot calculating operation of reflecting the shape of the steering wheel, which is pre-stored in the control unit, in the display module, and calculating an area of the display module, which is covered by the steering wheel and cannot be identified by the driver, by the control unit.

The utility model discloses a melon stock wood leaf display automatic identification device belongs to agricultural machinery technical field, including installation base, the upper surface of installation base is connected with fixed disc, the upper surface of fixed disc is installed with main motor, the upper surface of main motor is connected with installation cylinder, the top of installation cylinder is connected with feeding mechanism, the outer surface sliding of installation cylinder is connected with sliding ring, the outer surface of sliding ring is connected with two fixed plates, and the inside of two fixed plates all inlaying has rotation bearing, and the device whole course does not need manual intervention, and realizes automation closed loop from stock wood feeding to identification result output: feeding mechanism replaces manual plate arrangement, pneumatic clamping jaw replaces manual fixing, and pose adjustment and multi -camera identification replace manual overturn observation, and data processing unit replaces manual measurement and judgment, and then can significantly improve the work efficiency of batch identification, reduces manpower investment and labor intensity of manual operation simultaneously.

The utility model discloses a normal temperature container based on double-camera identification, which comprises a container body, a lamp box is arranged on the container body, the container body comprises a main body frame and a container door, and a left side container body plate, a right side container body plate and a back plate are arranged outside the main body frame; a plurality of laminates are arranged in the main body frame, and a first camera is arranged above the laminates; a main control box, an electronic lock, a second camera and a power supply are arranged in the lamp box, and a display screen is arranged outside the lamp box; the input end of the main control box is connected with the first camera and the second camera, and the output end of the main control box is connected with the electronic lock; and the main control box provides a two-dimensional code to the display screen and unlocks the electronic lock according to a code scanning result. According to the normal-temperature container, shopping can be conveniently achieved through full-dynamic camera recognition, the shopping process is simple, and the man-machine interaction degree is high; and a complicated mechanical structure is not needed, and the cabinet body is simple in assembly process, convenient to carry and convenient to operate.

The feeding device comprises a machine table, and a flexible vibration disc, a storage assembly, a moving mechanism and a conveying line are arranged on the machine table. The conveying line can drive the trays to the storage area; a suction mechanism is mounted on the moving mechanism and can be driven to move along an X axis and a Y axis; the material suction mechanism is composed of a lifting assembly, a vacuum suction pipe, a first recognition camera and a second recognition camera, the lifting assembly is used for driving the vacuum suction pipe to ascend and descend, the first recognition camera recognizes the position of a product in the flexible vibration disc, and the vacuum suction pipe is driven by the moving mechanism to grab the product and place the product to the storage assembly; the second recognition camera recognizes the position of the product on the containing assembly, the lifting assembly is connected with the rotating motor to drive the vacuum suction pipe to rotate, and after the vacuum suction pipe grabs the product again, the rotating motor conducts angle compensation on the product according to position information of the second recognition camera, and the product is accurately placed on the tray. According to the device, through the storage assembly and secondary recognition positioning, dynamic compensation adjustment of the angle of a product is achieved, and the feeding stability is improved.

The utility model relates to the technical field of semiconductor testing, and provides a probe assembly used for a flying probe testing machine and capable of identifying a needle point in the vertical direction, which comprises a main support, a cantilever support arranged on the main support, a probe arranged on the cantilever support, and a camera identification assembly arranged horizontally and used for identifying the position of the needle point of the probe in the vertical direction. The camera identification assembly comprises a mounting bracket arranged on the main bracket, an identification camera arranged on the mounting bracket, a lens matched with the identification camera and a prism matched with the lens; the camera identification assembly and the probe are always in a relative static state, the problem of interference and shielding of image acquisition does not exist, the needle point position of the probe in the vertical direction can be accurately and clearly identified through the camera identification assembly, so that debugging alignment operation is accurately carried out, the problem of mutual interference and influence does not exist among a plurality of structures, and the debugging precision is improved. Therefore, small bonding pad testing work can be carried out, and the application range of the flying probe testing machine is expanded.

1. Name of the product in this design: Camera Recognition Graphical User Interface for Electronic Devices. 2. Intended use of this design: for use in electronic devices. 3. The key design feature of this product is its graphical user interface. 4. The image or photo that best illustrates the design points: Design 1 Interface Change State Diagram 1. 5. The rear view, left view, right view, top view, and bottom view of this design are conventional designs, and the rear view, left view, right view, top view, and bottom view are omitted. 6. Design 1 is designated as the basic design. 7. Uses of graphical user interfaces: to enable recognition functions through the camera of electronic devices, such as translation functions. 8. Human-computer interaction method of graphical user interface: In Design 1 and Design 2, click the blue button at the bottom of the main view to switch to the interface change state diagram 1, enter the camera translation function, wait for a certain time in the interface change state diagram 1 or press the photo button on the electronic device to switch to the interface change state diagram 2.

The application provides an electric core aluminum shell assembling method, system, equipment and medium in the technical field of electric core production, and the method comprises the following steps: step S1, transmitting the shell and the aluminum cover through the PLC control conveying belt; step S2, the PLC returns the aluminum cover to the aluminum cover buffer area through the return flow line; step S3, the PLC identifies the buffer quantity of the shell buffer area and the aluminum cover buffer area, controls the work of the blocking cylinder of the conveying belt based on the buffer quantity, identifies the offset of the shell on the feeding belt line through the camera and transmits the offset to the cover assembling robot; step S4, the PLC controls the cover assembling robot, dynamically and automatically assembles the aluminum cover and the shell grabbed from the aluminum cover buffer area and the feeding belt line respectively, transmits the assembled aluminum shell to the aluminum shell buffer area, and grabs the aluminum shell to the rubber coating station by the grabbing robot; and step S5, the PLC records the assembling quantity of the aluminum shell and monitors the aluminum shell quantity in the assembling process in real time. The application has the advantages that the efficiency and quality of the electric core aluminum shell assembling are greatly improved.

A method for checking a possibility of inserting a component, according to an embodiment of the present invention, comprises the steps of: recognizing, by a first camera, the positions, sizes, and shapes of pin insertion portions of a PCB; storing the positions, sizes, and shapes of the pin insertion portions as data; checking, by a second camera, the positions of pins of a component to be inserted; adjusting the position of the component to be inserted so as to match the positions of the pins of the component to be inserted with the positions of the pin insertion portions, by using position data of the pin insertion portions; fixing the positions of the PCB and the pin insertion portions, recognizing the sizes and shapes of the pins of the component to be inserted, and storing same as data; and determining whether or not the pins of the component to be inserted can be inserted into the pin insertion portions, by matching the sizes and shapes of the pins of the component to be inserted with the pin insertion portions.

The invention relates to the technical field of sound image detection after cable plugging and unplugging of TV finished products, in particular to full-automatic hand-eye integrated power supply USB plug plugging equipment which comprises a belt conveying line, pallet clamp mechanisms are installed on the two sides of the top of the belt conveying line, and each pallet clamp mechanism comprises two conical guide pin clamping seats installed on the top of the belt conveying line. A limiting clamping groove is formed in the top of the conical guide pin clamping base, two sets of pins are installed on the two sides of the top of the conical guide pin clamping base, two magnetic mounting bases are connected into the limiting clamping groove in a clamped mode, and a limiting open groove is formed in one side of each magnetic mounting base; by adopting the magnetic mounting base on the large magnet groove for adsorption and fixation, compared with pin clamping, the difficulty that the clamps are placed into the clamp base is reduced, the plug clamp is suitable for occasions where wires on a pallet are not wound and pulled, the plug clamp is designed according to different appearance characteristics, and the plug clamp is simple in structure, convenient to use and high in practicability. And the relative position of the plug and the clamp can be floated, so that damage to the TV terminal during rigid plugging and unplugging is avoided, and accurate positioning of the USB floating plug clamp is realized.

An automatic laser lofting system based on a mobile robot platform, comprising a mobile platform, a mechanical arm, an end effector, a control box and a three-dimensional laser projection positioning system, wherein the mechanical arm is movably mounted on the mobile platform; the end effector is carried at the end of the mechanical arm; the three-dimensional laser projection positioning system manipulates the end effector to dynamically project lofting points to the construction site to present accurate visible laser points for 2D industrial camera recognition; the control box uniformly coordinates the actions of the mobile platform, the mechanical arm, the end effector and the three-dimensional laser projection positioning system.

The invention discloses an AGV scheduling management system based on multi-line sensing, which belongs to the technical field of artificial intelligence systems, is used for AGV scheduling management, and comprises a control module, a scheduling module, a storage module and a vision module, when an automatic task is started, a full barrel or an empty barrel is found through the vision module, and when a proper position is found, the storage module generates an automatic task, and the automatic task is sent to the control module; and the scheduling plate plans a walking route, distributes vehicles and notifies the control plate, and the control program and the vehicles interact to complete a command issued by scheduling. An internal thread of the control plate comprises a vehicle data connection and a vehicle dispatcher; the scheduling plate performs real-time positioning and updating; an internal thread of the storage module comprises a task planner and a storage service; the internal thread of the visual module comprises reasoning and data synchronization, the reasoning is real-time camera recognition, and the data synchronization is recognition state updating. Compared with the prior art, the AGV operation efficiency is improved, task allocation is reasonably regulated and controlled, and an efficient, intelligent and flexible modern AGV system is constructed.

The invention discloses a self-service vending method and a self-service vending machine, and the method comprises the steps: inputting the types and prices of commodities put into the self-service vending machine based on a database module to establish a commodity database, and displaying the types and prices of all commodities in the self-service vending machine on a display interface of the self-service vending machine based on a man-machine interaction module, commodity taken out from the self-service vending machine by a user is recognized based on the camera recognition module, if the user doubts the category or price of the self-purchased commodity, the user can perform secondary confirmation and selection of the commodity based on the man-machine interaction module, and a two-dimensional code on the commodity can be scanned based on the commodity code scanning module to confirm the price of the commodity. The category and the price of the commodity are displayed on the basis of the display interface of the man-machine interaction module again, so that the user can confirm the self-purchased commodity again; according to the scheme, the commodities selected by the user in the self-service vending machine can be subjected to secondary code scanning confirmation, and errors of automatically deducting the amount of the commodities of the user due to system errors or mistaken recognition of the commodities by a camera are effectively avoided.

The invention discloses a cut tobacco flow optimization control method before cut tobacco drying. Sufficient material supply in subsequent regulation and control is guaranteed by adjusting the material distribution condition in the storage cabinet; the starting and stopping times, namely the material supply times, of equipment in a period of time are adopted as a detection point for reducing the storage cabinet bottom band frequency, the low material level of a metering grating is adopted as a detection point for increasing the storage cabinet bottom band frequency, the cabinet outlet frequency is continuously increased and reduced through the two detection points, and an optimal bottom band frequency mean value is achieved; a 3D industrial line scanning camera is erected to recognize the change condition of materials in the storage cabinet, a control model is established to enable a storage cabinet bottom belt and a lifting belt to adjust the discharging frequency in a self-adaptive mode, accurate control over the storage cabinet is achieved, the starting and stopping frequency of equipment is greatly reduced, and materials are stably supplied. Meanwhile, tobacco shred breakage caused by too many starting and stopping times is reduced, the load and abrasion of the equipment are large when the equipment is started, the maintenance cost of the equipment and the maintenance workload of personnel are reduced due to reduction of the starting and stopping times, and the service life of the equipment is prolonged.

The utility model relates to the technical field of parking management, and discloses a parking lot off-line payment system, which comprises a payment server, a camera, a payment two-dimensional code display screen and a code scanning device, and is characterized in that the camera is used for identifying license plate information of a leaving vehicle and sending the license plate information to the payment server; the payment service determines the parking fee information of the departing vehicle by using the license plate information identified by the camera, and generates an offline payment two-dimensional code when determining that the connection with the cloud service platform is in an offline state of network interruption; the payment two-dimensional code display screen receives and displays the offline payment two-dimensional code; the code scanning device identifies a successful payment two-dimensional code displayed on the mobile terminal after the vehicle owner scans the offline payment two-dimensional code through the mobile terminal to pay the parking fee; and the payment server is also used for verifying the payment success two-dimensional code information, and after the verification is passed, the barrier gate is controlled to automatically lift the rod for release. Therefore, the response capability of the parking lot under the network disconnection is improved, the normal operation of the parking lot under the network disconnection is realized, and the parking experience and satisfaction of the vehicle owner are also improved.

A LiDAR lane detection method based on a deep segmentation model uses point cloud data collected by intelligent driving devices during driving and employs a multi-scale deep convolutional network to detect lane markings. This method uses deep learning technology in conjunction with the intelligent driving device's LiDAR to iteratively update the model. This method, unlike traditional camera-based recognition methods, reduces the hardware requirements of intelligent driving devices and reduces their cost.

The invention provides a robot positioning method and system based on multi-sensor fusion and a semantic map. The method comprises the steps that a monocular camera is used for recognizing a semantic object, and the direction angle of the semantic object in a camera coordinate system is calculated; the laser radar point cloud is converted to a camera coordinate system, and accurate position information of the object is acquired in combination with semantic direction information; constructing an environment map with semantic annotation as priori knowledge; the current pose is calculated through semantic matching during navigation and compared with a system positioning result, and when the pose error is smaller than a threshold value, it is considered that robot positioning is correct; and otherwise, positioning again and recovering accurate positioning by taking the semantic calculation pose as a reference. The method has the advantages of low hardware cost, high environmental adaptability and the like, the positioning problem caused by dynamic interference and feature deficiency is effectively solved, the positioning precision and robustness of the robot in a complex environment are remarkably improved, and the method can be widely applied to a positioning navigation system of the robot.

An image processing system includes: an image acquisition unit configured to acquire image data generated by an imaging device that captures an optical image having a low-distortion region and a high-distortion region; a first image recognition unit configured to perform image recognition on image data of at least a partial region in the image data acquired by the image acquisition unit to output a first image recognition result; a second image recognition unit configured to perform image recognition on image data of a region wider than the partial region in the image data acquired by the image acquisition unit to output a second image recognition result; and an integration processing unit configured to output an image recognition result integrated on the basis of the first image recognition result and the second image recognition result.

The application discloses a dust-removing and color-detecting device, which comprises a machine body and a detection box, a control system is arranged in the machine body, a dust-removing mechanism and a color-detecting mechanism are arranged in the detection box, the control system is connected with the dust-removing mechanism and the color-detecting mechanism respectively, a visual detection assembly comprises an identification camera and a detection camera, and a light assembly comprises a plurality of lamps which emit different color lights respectively. The detection method of the dust-removing and color-detecting device, the control system controls the dust-removing mechanism to perform dust-removing treatment on a product to be detected, then controls the identification camera to identify the model of the product to be detected, thereby controlling corresponding lamps to emit light, and the detection camera performs color-difference detection on the product. The dust-removing and color-detecting device and the detection method thereof are suitable for color-difference detection requirements of different products by arranging lamps which emit different color lights, and automatic identification and light switching are realized by cooperation of the control system and the identification camera, so that complicated steps of manually replacing lamps are omitted, operation is simple and convenient, and detection efficiency is effectively improved.