CN118408185B - Agricultural heat dissipation type three-proofing lighting lamp and manufacturing method thereof - Google Patents

Abstract

The invention discloses an agricultural heat dissipation type three-proofing lighting lamp and a preparation method thereof. The lighting fixture comprises: a metal rod body having a shape space; the laser diode is electrically connected to the driving circuit board, the heat dissipation plate is attached to the back of the driving circuit board, the lens system is kept on the light emitting path of the laser diode, and the protective glass is arranged above the lens system; and a heat sink extending outward from the metal rod. The structure setting makes the three-proofing light of agricultural heat dissipation type can be applicable to various abominable agricultural environment.

Description

Agricultural heat dissipation type three-proof lighting fixture and preparation method thereof

Technical Field

The present application relates to the field of lighting fixtures, and more specifically, to an agricultural heat dissipation type three-proof lighting fixture and a preparation method thereof.

Background Art

With the development of agricultural modernization, agricultural lighting plays a vital role in the fields of planting, breeding, etc. For example, lighting fixtures can create a suitable light environment through artificial light, or make up for the lack of natural light, regulate the growth and reproduction of agricultural organisms, and thus improve production efficiency and output.

Traditional lighting equipment is prone to problems such as heating and poor heat dissipation after long-term operation, which affects the life and stability of the equipment. In addition, some lamps have problems in the manufacturing process, which may lead to quality defects such as unstable parts and loose connections, affecting the use effect and safety of the lamps.

Therefore, an optimized agricultural heat dissipation type three-proof lighting fixture and a preparation method thereof are expected.

Summary of the invention

In view of this, the present application proposes an agricultural heat dissipation type triple-proof lighting fixture and a preparation method thereof.

According to one aspect of the present application, there is provided an agricultural heat dissipation type three-proof lighting fixture, comprising:

A metal rod body having a shape space;

A laser diode, a driving circuit board, a heat sink, a lens system and a protective glass installed in the shape space, wherein the laser diode is electrically connected to the driving circuit board, the heat sink is attached to the back of the driving circuit board, the lens system is held on the light output path of the laser diode, and the protective glass is arranged above the lens system; and

A heat sink extends outward from the metal rod.

In the above-mentioned agricultural heat dissipation type three-proof lighting fixture, the metal rod body is a 304 stainless steel rod.

In the above-mentioned agricultural heat dissipation type three-proof lighting fixture, the heat sink is in the shape of petals.

According to another aspect of the present application, a method for preparing an agricultural heat dissipation type triple-proof lighting fixture is provided, comprising:

Provide stainless steel rods, laser diodes, driver circuit boards, heat sinks, lens systems, and protective glass;

Assembling the laser diode, the driving circuit board and the heat sink to obtain a light-emitting assembly;

Processing the stainless steel rod by laser cutting so that the stainless steel rod has a desired shape space;

Installing the light-emitting assembly in the shape space of the stainless steel rod, and installing the lens system and the protective glass to obtain a light-emitting body; and

The light-emitting body is installed into the petal-shaped heat sink.

In the above-mentioned method for preparing the agricultural heat dissipation type three-proof lighting fixture, the stainless steel rod is processed by laser cutting so that the stainless steel rod has a desired shape space, including:

The camera collects the cutting state image of the stainless steel bar after laser cutting;

Extracting reference post-cutting state images marked as qualified from a background database;

Performing image feature extraction on the cut-out state image and the reference cut-out state image marked as qualified to obtain a sequence of cut-out state image block feature vectors and a sequence of reference cut-out image block feature vectors; and

The quality inspection result is determined based on global difference degree information between the sequence of feature vectors of the image block in the segmented state and the sequence of feature vectors of the image block in the reference segmented state.

In the above-mentioned method for preparing the agricultural heat dissipation type three-proof lighting fixture, image feature extraction is performed on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image block feature vectors and a sequence of reference cut state image block feature vectors, including:

Performing image preprocessing on the cut-out state image and the reference cut-out state image marked as qualified to obtain a sequence of cut-out state image blocks and a sequence of reference cut-out image blocks respectively; and

The sequence of the cut state image blocks and the sequence of the reference cut image blocks are respectively passed through a feature extractor based on a convolutional neural network model to obtain a sequence of feature vectors of the cut state image blocks and a sequence of feature vectors of the reference cut image blocks.

In the above-mentioned method for preparing the agricultural heat dissipation type three-proof lighting fixture, image preprocessing is performed on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image blocks and a sequence of reference cut state image blocks, respectively, including:

The cut-out state image and the reference cut-out state image marked as qualified are respectively divided into image blocks to obtain a sequence of the cut-out state image blocks and a sequence of the reference cut-out image blocks.

In the above-mentioned method for preparing the agricultural heat dissipation type three-proof lighting fixture, the quality inspection result is determined based on the global difference degree information between the sequence of the feature vectors of the image block after cutting and the sequence of the feature vectors of the reference image block after cutting, including:

Respectively calculating the image local feature difference semantic metric coefficients between each group of corresponding post-segmentation state image block feature vectors and reference post-segmentation image block feature vectors in the sequence of post-segmentation state image block feature vectors and the sequence of reference post-segmentation image block feature vectors to obtain a global semantic difference representation feature vector composed of multiple image local feature difference semantic metric coefficients;

Performing feature distribution clustering optimization on the global semantic difference representation feature vector to obtain an optimized global semantic difference representation feature vector; and

The optimized global semantic difference characterization feature vector is passed through a quality inspector based on a classifier to obtain a quality inspection result, and the quality inspection result is used to indicate whether it is qualified.

In the above-mentioned method for preparing the agricultural heat dissipation type three-proof lighting fixture, the image local feature difference semantic metric coefficients between each group of corresponding cut state image block feature vectors and reference cut image block feature vectors in the sequence of the cut state image block feature vectors and the sequence of the reference cut image block feature vectors are calculated respectively to obtain a global semantic difference representation feature vector composed of multiple image local feature difference semantic metric coefficients, including:

The image local feature difference semantic measurement coefficient between each group of corresponding post-cut state image block feature vectors and reference post-cut image block feature vectors in the sequence of the post-cut state image block feature vectors and the sequence of the reference post-cut image block feature vectors is calculated by the following image local feature difference measurement formula; wherein the image local feature difference measurement formula is: ;in, is the first in the sequence of feature vectors of the image block after cutting. The feature vector of the image block after cutting, is the first in the sequence of the reference cut image block feature vectors reference image block feature vector after cutting, For the The semantic measurement coefficient of the local feature difference of the image, represents the square of the L2 norm of the eigenvector, Indicates the bitwise subtraction of elements.

In the present application, the lighting fixture includes: a metal rod body with a shape space; a laser diode, a driving circuit board, a heat sink, a lens system and a protective glass installed in the shape space, wherein the laser diode is electrically connected to the driving circuit board, the heat sink is attached to the back of the driving circuit board, the lens system is held on the light output path of the laser diode, and the protective glass is arranged above the lens system; and a heat sink extending outward from the metal rod body. Such a structural setting enables the agricultural heat dissipation type three-proof lighting lamp to be suitable for various harsh agricultural environments.

Other features and aspects of the present application will become apparent from the following detailed description of the present application with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the present application and, together with the description, serve to explain the principles of the present application.

FIG1 is a schematic structural diagram of a metal rod according to an embodiment of the present application.

FIG2 shows a flow chart of a method for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

FIG3 shows a flow chart of sub-step S130 of the method for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

FIG. 4 shows a flow chart of sub-step S133 of the method for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

FIG5 shows a flow chart of sub-step S134 of the method for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

FIG6 shows a block diagram of a system for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

FIG. 7 shows an application scenario diagram of a method for preparing an agricultural heat dissipation type triple-proof lighting fixture according to an embodiment of the present application.

DETAILED DESCRIPTION

The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and/or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

In addition, in order to better illustrate the present application, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present application can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present application.

The present application provides an agricultural heat dissipation type three-proof lighting fixture, which specifically includes: a metal rod body with a shape space; a laser diode, a driving circuit board, a heat sink, a lens system and a protective glass installed in the shape space, wherein the laser diode is electrically connected to the driving circuit board, the heat sink is attached to the back of the driving circuit board, the lens system is maintained on the light output path of the laser diode, and the protective glass is arranged above the lens system; and a heat sink extending outward from the metal rod body. Among them, the structure of the metal rod body is shown in Figure 1. In a specific example of the present application, the metal rod body is a 304 stainless steel rod, and the heat sink is in the shape of a petal. More specifically, the agricultural heat dissipation type three-proof lighting fixture uses a metal rod body as a main bracket, and a laser diode, a driving circuit board, a heat sink, a lens system and a protective glass are installed inside, which can improve the heat dissipation effect and stability of the lamp. At the same time, a heat sink is arranged at the outward extension of the metal rod body to further increase the heat dissipation area, which is conducive to the improvement of the heat dissipation effect. Such a structural setting makes the agricultural heat dissipation type three-proof lighting fixture suitable for various harsh agricultural environments.

In particular, the present application also provides a method for preparing an agricultural heat dissipation type three-proof lighting fixture, and FIG2 shows a flow chart of the method for preparing an agricultural heat dissipation type three-proof lighting fixture according to an embodiment of the present application. As shown in FIG2, the method for preparing an agricultural heat dissipation type three-proof lighting fixture according to an embodiment of the present application includes the following steps: S110, providing a stainless steel rod, a laser diode, a driving circuit board, a heat dissipation plate, a lens system and a protective glass; S120, assembling the laser diode, the driving circuit board and the heat dissipation plate to obtain a light-emitting component; S130, processing the stainless steel rod by laser cutting so that the stainless steel rod has a desired shape space; S140, installing the light-emitting component in the shape space of the stainless steel rod, and installing the lens system and the protective glass to obtain a light-emitting body; and, S150, installing the light-emitting body into a petal-shaped heat sink.

Among them, in the actual preparation process of the agricultural heat dissipation type three-proof lighting fixture, when the stainless steel rod is processed by laser cutting so that the stainless steel rod has the required shape space, it is usually necessary to conduct quality inspection on the cut stainless steel rod. The importance of quality inspection lies in ensuring that the cut stainless steel rod meets the design requirements and has the required quality. Specifically, although laser cutting is precise, there is still a certain cutting error. Quality inspection can ensure that the dimensional accuracy of the stainless steel rod meets the design specifications to ensure the effect of subsequent assembly. At the same time, it is taken into account that the cutting process may affect the surface quality of the stainless steel rod, such as producing burrs, cracks and other defects. These defective stainless steel rods can be eliminated through quality inspection to avoid affecting the appearance and performance of the final product.

The existing quality inspection method is usually carried out manually, for example, the surface quality of the cut stainless steel bar is visually inspected by technicians to observe whether there are defects such as burrs, cracks, deformation, etc., and the size of the cut stainless steel bar is measured by technicians manually to ensure that its size meets the design requirements. However, this method of manual visual inspection and manual measurement is easily affected by human subjective factors and may lead to misjudgment or error. In addition, this method usually requires a lot of manpower and time, and is inefficient. Therefore, an optimized solution is expected.

In response to the above technical problems, the technical concept of the present application is: introducing a reference cutting state image marked as qualified, and combining it with an intelligent algorithm to extract and mine image features of the reference cutting state image of the stainless steel bar after laser cutting, and then performing differentiation measurement in the feature space to characterize the degree of image semantic difference between the two, and based on this image semantic difference degree, intelligently judge whether the stainless steel bar after laser cutting is qualified.

Based on this, as shown in FIG3 , in step S130, the stainless steel rod is processed by laser cutting so that the stainless steel rod has a desired shape space, including: S131, collecting a post-cutting state image of the stainless steel rod after laser cutting by a camera; S132, extracting a reference post-cutting state image marked as qualified from a background database; S133, performing image feature extraction on the post-cutting state image and the reference post-cutting state image marked as qualified to obtain a sequence of post-cutting state image block feature vectors and a sequence of reference post-cutting image block feature vectors; and, S134, determining a quality inspection result based on global difference degree information between the sequence of post-cutting state image block feature vectors and the sequence of reference post-cutting image block feature vectors.

Specifically, when the stainless steel bar is processed by laser cutting so that the stainless steel bar has the required shape space, it also includes: first, collecting the post-cutting state image of the stainless steel bar after laser cutting by a camera; and extracting the reference post-cutting state image marked as qualified from the background database. Here, collecting the post-cutting state image of the stainless steel bar after laser cutting by a camera can record and reflect the state information and cutting effect of the stainless steel bar. Among them, the reference post-cutting state image is a sample that has been approved and marked as qualified. The reference post-cutting state image represents the ideal post-cutting state, which meets the product design requirements and quality standards. Using the reference post-cutting state image as a reference can help determine whether the post-cutting state image meets the expected standards.

Then, the cut state image and the reference cut state image marked as qualified are respectively segmented into image blocks to obtain a sequence of cut state image blocks and a sequence of reference cut state image blocks. Here, the cut state image and the reference cut state image marked as qualified are segmented into image blocks respectively, which can guide the model to perform more accurate feature analysis on each image block to capture subtle changes and detail information contained in each image block, which is helpful to find local defects or differences during the quality inspection process.

Next, the sequence of the cut state image blocks and the sequence of the reference cut image blocks are respectively passed through a feature extractor based on a convolutional neural network model to obtain a sequence of feature vectors of the cut state image blocks and a sequence of feature vectors of the reference cut image blocks. Among them, the convolutional neural network (CNN) is a deep learning model specifically used to process data with a grid structure, such as images. Specifically, the CNN model can effectively learn abstract feature representations in images through components such as convolutional layers, pooling layers, and activation layers. In the technical solution of the present application, the feature extractor is constructed by utilizing the excellent performance of the convolutional neural network model in image feature extraction to capture the image semantic features and local neighborhood spatial association patterns of each of the cut state image blocks and each of the reference cut image blocks, respectively.

Correspondingly, as shown in FIG4 , in step S133, image feature extraction is performed on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image block feature vectors and a sequence of reference cut image block feature vectors, including: S1331, performing image preprocessing on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image blocks and a sequence of reference cut image blocks; and, S1332, passing the sequence of cut state image blocks and the sequence of reference cut image blocks through a feature extractor based on a convolutional neural network model to obtain a sequence of cut state image block feature vectors and a sequence of reference cut image block feature vectors.

It should be understood that in step S1331, the post-cut state image and the reference post-cut state image are pre-processed to extract image blocks, which is used to: adjust the image size or resolution to meet the requirements of the feature extractor; divide the image into overlapping or non-overlapping image blocks. In step S1332, a convolutional neural network model is used to extract features from the image blocks, which is used to: apply the convolutional neural network model to extract features related to the post-cut state in the image blocks, such as texture, shape and edge, and generate a sequence of post-cut state image block feature vectors and a sequence of reference post-cut image block feature vectors.

Wherein, in step S1331, the cut state image and the reference cut state image marked as qualified are respectively subjected to image preprocessing to obtain a sequence of cut state image blocks and a sequence of reference cut image blocks, including: performing image block segmentation on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image blocks and a sequence of reference cut image blocks.

It should be understood that the purpose of step S1331 is to divide the image into overlapping or non-overlapping image blocks. It can allow the feature extractor to focus on specific areas of the image, thereby reducing computational costs, because the feature extractor can process image blocks in parallel, improving the robustness of the features, because local changes in image blocks are less likely to affect global features. And for some types of feature extractors, such as feature extractors based on convolutional neural networks, image block segmentation is necessary because these models require fixed-size inputs. In the analysis of the post-cut state image, image block segmentation can: capture local features of the post-cut state, such as texture, shape, and edges; allow the feature extractor to identify different aspects of the post-cut state in the image; improve the discriminability of the features, thereby improving the classification or evaluation of the post-cut state.

Subsequently, the image local feature difference semantic metric coefficients between each group of corresponding image block feature vectors and reference image block feature vectors in the sequence of the image block feature vectors after cutting and the sequence of the image block feature vectors after cutting are calculated to obtain a global semantic difference representation feature vector composed of multiple image local feature difference semantic metric coefficients. That is, the image difference degree between the local image blocks in the image after cutting and the reference image after cutting is comprehensively evaluated by calculating the image local feature difference semantic metric coefficients between each group of corresponding image block feature vectors to quantify the difference between the image after cutting and the reference image after cutting, and provide an important basis for determining whether the stainless steel bar in the image after cutting is qualified. In particular, the global semantic difference representation feature vector composed of the semantic metric coefficients of the local feature difference of each image can more comprehensively describe the overall semantic difference between the image after cutting and the reference image after cutting. In other words, this global representation information can characterize and describe the overall relationship between the image blocks, which helps the system to more accurately judge the overall difference degree between the stainless steel bar after cutting and the reference stainless steel bar marked as qualified. Then, the global semantic difference characterization feature vector is passed through a quality checker based on a classifier to obtain a quality check result, and the quality check result is used to indicate whether it is qualified.

Correspondingly, as shown in FIG5 , in step S134, the quality inspection result is determined based on the global difference degree information between the sequence of the feature vectors of the image blocks in the cut state and the sequence of the feature vectors of the reference image blocks in the cut state, including: S1341, respectively calculating the image local feature difference semantic measurement coefficients between each group of corresponding feature vectors of the image blocks in the cut state and the sequence of the feature vectors of the reference image blocks in the cut state to obtain a global semantic difference representation feature vector composed of multiple image local feature difference semantic measurement coefficients; S1342, performing feature distribution clustering optimization on the global semantic difference representation feature vector to obtain an optimized global semantic difference representation feature vector; and, S1343, passing the optimized global semantic difference representation feature vector through a quality inspector based on a classifier to obtain a quality inspection result, and the quality inspection result is used to indicate whether it is qualified.

Wherein, in step S1341, respectively calculating the image local feature difference semantic measurement coefficients between each group of corresponding post-cut state image block feature vectors and reference post-cut image block feature vectors in the sequence of the post-cut state image block feature vectors and the sequence of the reference post-cut image block feature vectors to obtain a global semantic difference representation feature vector composed of multiple image local feature difference semantic measurement coefficients, including: respectively calculating the image local feature difference semantic measurement coefficients between each group of corresponding post-cut state image block feature vectors and reference post-cut image block feature vectors in the sequence of the post-cut state image block feature vectors and the sequence of the reference post-cut image block feature vectors using the following image local feature difference measurement formula; wherein, the image local feature difference measurement formula is: ;in, is the first in the sequence of feature vectors of the image block after cutting. The feature vector of the image block after cutting, is the first in the sequence of the reference cut image block feature vectors reference image block feature vector after cutting, For the The semantic measurement coefficient of the local feature difference of the image, represents the square of the L2 norm of the eigenvector, Indicates the bitwise subtraction of elements.

In the above technical solution, the sequence of the feature vectors of the cut state image blocks and the sequence of the feature vectors of the reference cut state image blocks respectively express the image semantic features of the cut state image and the reference cut state image in the local image semantic space domain. However, considering the uneven spatial distribution of the image source semantics in the local image semantic space domain in the global image semantic space domain, each group of corresponding cut state image block feature vectors and reference cut state image block feature vectors in the sequence of the cut state image block feature vectors and the sequence of the reference cut state image block feature vectors will also have image semantic feature distribution non-correspondence, so that the global semantic difference representation feature vector composed of the image local feature difference semantic measurement coefficients between each group of corresponding cut state image block feature vectors and reference cut state image block feature vectors will have significant local feature distribution discreteness.

In this way, when the global semantic difference characterization feature vector is used as a whole for classification and regression through a classifier, the discreteness of the local feature distribution of the global semantic difference characterization feature vector will lead to difficulty in converging towards the probability of a predetermined class when the classification and regression is performed through the classifier, thereby affecting the training speed of the classifier and the accuracy of the final classification result.

Based on this, the applicant of the present application performs clustering optimization on the global semantic difference representation feature vector, that is, first clustering the eigenvalues of the global semantic difference representation feature vector, for example, clustering based on the distance between eigenvalues, and then optimizing the in-class and out-of-class representations based on the clustered features.

Accordingly, in one example, in step S1342, the global semantic difference characterization feature vector is subjected to feature distribution clustering optimization to obtain an optimized global semantic difference characterization feature vector, including: performing feature distribution clustering optimization on the global semantic difference characterization feature vector according to the following optimization formula to obtain the optimized global semantic difference characterization feature vector; wherein the optimization formula is: ;in, are the eigenvalues of the global semantic difference characterization feature vector, is the number of feature sets corresponding to the global semantic difference characterization feature vector, that is, the length of the global semantic difference characterization feature vector, is the number of clustering features, represents the clustering feature set, are the eigenvalues of the optimized global semantic difference representation feature vector.

Specifically, by taking the in-class features and out-of-class features of the global semantic difference characterization feature vector as different instance roles to perform class instance description based on clustering proportion distribution, and introducing the clustering response history based on the in-class and out-of-class dynamic context, a coordinated global perspective is maintained on the in-class distribution and out-of-class distribution of the overall features of the global semantic difference characterization feature vector, so that the optimized feature clustering operation of the global semantic difference characterization feature vector can maintain a consistent response of the in-class and out-of-class features, so that the regression convergence path based on feature clustering remains consistent during the class regression process, and the convergence effect of the global semantic difference characterization feature vector towards the predetermined class probability is improved, so as to improve the training speed of the classifier and the accuracy of the classification results.

Further, in step S1343, the optimized global semantic difference representation feature vector is passed through a classifier-based quality inspector to obtain a quality inspection result, and the quality inspection result is used to indicate whether it is qualified, including: using the fully connected layer of the classifier-based quality inspector to fully connect encode the optimized global semantic difference representation feature vector to obtain an encoded classification feature vector; and inputting the encoded classification feature vector into the Softmax classification function of the classifier-based quality inspector to obtain the quality inspection result.

It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

In summary, based on the preparation method of the agricultural heat dissipation type three-proof lighting fixture of the embodiment of the present application, it can intelligently determine whether the stainless steel rod after laser cutting is qualified.

FIG6 shows a block diagram of a system 100 for preparing agricultural heat dissipation type three-proof lighting fixtures according to an embodiment of the present application. As shown in FIG6, the system 100 for preparing agricultural heat dissipation type three-proof lighting fixtures according to an embodiment of the present application includes: a cutting image acquisition module 110, which is used to collect the cut state image of the stainless steel bar after laser cutting through a camera; a reference image acquisition module 120, which is used to extract the reference cut state image marked as qualified from the background database; an image feature extraction module 130, which is used to extract image features from the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image block feature vectors and a sequence of reference cut image block feature vectors; and a quality inspection analysis module 140, which is used to determine the quality inspection result based on the global difference degree information between the sequence of the cut state image block feature vectors and the sequence of the reference cut image block feature vectors.

Here, those skilled in the art can understand that the specific functions and operations of each unit and module in the above-mentioned agricultural heat dissipation type three-proof lighting fixture preparation system 100 have been described in detail in the description of the preparation method of the agricultural heat dissipation type three-proof lighting fixture with reference to Figures 2 to 5 above, and therefore, its repeated description will be omitted.

As described above, the preparation system 100 of agricultural heat dissipation type three-proof lighting fixtures according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a preparation algorithm of agricultural heat dissipation type three-proof lighting fixtures. In a possible implementation, the preparation system 100 of agricultural heat dissipation type three-proof lighting fixtures according to the embodiment of the present application can be integrated into the wireless terminal as a software module and/or a hardware module. For example, the preparation system 100 of agricultural heat dissipation type three-proof lighting fixtures can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the preparation system 100 of agricultural heat dissipation type three-proof lighting fixtures can also be one of the many hardware modules of the wireless terminal.

Alternatively, in another example, the preparation system 100 of the agricultural heat-dissipating tri-proof lighting fixture and the wireless terminal may also be separate devices, and the preparation system 100 of the agricultural heat-dissipating tri-proof lighting fixture may be connected to the wireless terminal via a wired and/or wireless network, and transmit interactive information in accordance with an agreed data format.

FIG7 shows an application scenario diagram of the method for preparing agricultural heat dissipation type three-proof lighting fixtures according to an embodiment of the present application. As shown in FIG7, in this application scenario, first, a cutting state image of a stainless steel bar after laser cutting (for example, D1 shown in FIG7) is collected by a camera, and a reference cutting state image marked as qualified is extracted from a background database (for example, D2 shown in FIG7), and then the cutting state image and the reference cutting state image marked as qualified are input into a server (for example, S shown in FIG7) deployed with a preparation algorithm for agricultural heat dissipation type three-proof lighting fixtures, wherein the server can use the preparation algorithm for agricultural heat dissipation type three-proof lighting fixtures to process the cutting state image and the reference cutting state image marked as qualified to obtain a quality inspection result indicating whether the cutting state image and the reference cutting state image marked as qualified.

In an exemplary embodiment, there is also provided a non-volatile computer-readable storage medium, such as a memory including computer program instructions, which can be executed by a processing component of an apparatus to perform the above method.

The present application may be a system, a method and/or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present application.

Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive list) of computer readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination thereof. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (for example, a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the function involved. It should also be noted that each square box in the block diagram and/or flow chart, and the combination of the square boxes in the block diagram and/or flow chart can be realized by a dedicated hardware-based system that performs the function or action of the specification, or can be realized by a combination of special-purpose hardware and computer instructions.

The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims (4)

1. The preparation method of the agricultural heat dissipation type three-proofing lighting lamp is characterized by comprising the following steps of:

Providing a stainless steel rod, a laser diode, a driving circuit board, a heat dissipation plate, a lens system and protective glass;

assembling the laser diode, the driving circuit board and the heat dissipation plate to obtain a light emitting assembly;

Machining the stainless steel rod by laser cutting so that the stainless steel rod has a desired shape space;

The luminous component is arranged in the shape space of the stainless steel rod, and the lens system and the protective glass are arranged to obtain a luminous main body; and

Mounting the light-emitting body into a petal-shaped heat sink;

wherein the stainless steel rod is processed by laser cutting so that the stainless steel rod has a desired shape space, comprising:

collecting a cut state image of the laser cut stainless steel rod through a camera;

Extracting a reference cut state image marked as qualified from a background database;

extracting image features of the cut state image and the reference cut state image marked as qualified to obtain a sequence of feature vectors of the cut state image block and a sequence of feature vectors of the reference cut image block; and

Determining a quality inspection result based on global degree of difference information between the sequence of cut state image block feature vectors and the sequence of reference cut state image block feature vectors;

Wherein determining a quality inspection result based on global degree of difference information between the sequence of cut state image block feature vectors and the sequence of reference cut image block feature vectors comprises:

Respectively calculating image local feature difference semantic measurement coefficients between each group of corresponding cut state image block feature vectors and reference cut state image block feature vectors in the sequence of the cut state image block feature vectors and the sequence of the reference cut state image block feature vectors to obtain a global semantic difference characterization feature vector composed of a plurality of image local feature difference semantic measurement coefficients;

performing feature distribution cluster optimization on the global semantic difference characterization feature vector to obtain an optimized global semantic difference characterization feature vector; and

The optimized global semantic difference characterization feature vector passes through a quality detector based on a classifier to obtain a quality detection result, wherein the quality detection result is used for indicating whether the quality detection result is qualified or not;

performing feature distribution cluster optimization on the global semantic difference characterization feature vector to obtain an optimized global semantic difference characterization feature vector, including: performing feature distribution cluster optimization on the global semantic difference characterization feature vector by using the following optimization formula to obtain the optimized global semantic difference characterization feature vector; wherein, the optimization formula is: ; wherein, Is the respective feature value of the global semantic difference characterization feature vector,Is the number of feature sets corresponding to the global semantic difference characterization feature vector, i.e. the length of the global semantic difference characterization feature vector,Is the number of cluster features,A set of cluster features is represented,Is the respective feature value of the optimized global semantic difference characterization feature vector.

2. The method for manufacturing the agricultural heat dissipation type tri-proof lighting lamp according to claim 1, wherein the step of extracting image features of the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image block feature vectors and a sequence of reference cut image block feature vectors comprises:

Respectively carrying out image preprocessing on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image blocks and a sequence of reference cut image blocks; and

And respectively passing the sequence of the cut state image blocks and the sequence of the reference cut image blocks through a feature extractor based on a convolutional neural network model to obtain the sequence of the feature vectors of the cut state image blocks and the sequence of the feature vectors of the reference cut image blocks.

3. The method for manufacturing an agricultural heat dissipation type tri-proof lighting lamp according to claim 2, wherein the image preprocessing is performed on the cut state image and the reference cut state image marked as qualified to obtain a sequence of cut state image blocks and a sequence of reference cut state image blocks, respectively, comprising:

And respectively carrying out image block segmentation on the cut state image and the reference cut state image marked as qualified so as to obtain a sequence of the cut state image blocks and a sequence of the reference cut image blocks.

4. The method for manufacturing an agricultural heat dissipation type tri-proof lighting fixture according to claim 3, wherein calculating the image local feature difference semantic measurement coefficients between each set of corresponding cut state image block feature vectors and reference cut image block feature vectors in the sequence of cut state image block feature vectors and the sequence of reference cut image block feature vectors to obtain a global semantic difference characterization feature vector composed of a plurality of image local feature difference semantic measurement coefficients, respectively, comprises:

calculating the image local feature difference semantic measurement coefficients between each group of corresponding cut state image block feature vectors and reference cut image block feature vectors in the sequence of cut state image block feature vectors and the sequence of reference cut image block feature vectors respectively according to the following image local feature difference measurement formulas; the image local characteristic difference measurement formula is as follows: ; wherein, In the sequence of feature vectors for the cut state image blockThe feature vectors of the state image blocks after the cutting,In the sequence of feature vectors for the reference cut image blockThe reference cut image block feature vectors,Is the firstEach of said image local feature difference semantic metric coefficients,Representing the square of the L2 norm of the feature vector,Representing element pair bit subtraction difference processing.