30results about How to "Increase the number of operations" patented technology

The invention provides a stereo matching optimization method of a binocular visual system. The method comprises a step of using the binocular visual system to gain a binocular image; a step of correcting a distortion and polar line of the binocular visual system; a step of transferring the binocular image from a colorful image into a grayscale image; a step of regarding one of the grayscale images as a basic standard and matching with another image according to pixel masterplate; a step of optimizing ZNCC factor; a step of regarding the optimized ZNCC as a similarity measure to calculate in a matched mode so as to ensure matching parallax of each pixel point; a step of producing a dense disparity map according to the matching parallax of each pixel point. The method has the advantages of reducing times of calculation greatly and reducing calculating load of computer chip greatly by optimizing a computational formula of the ZNCC factors and calculating method. The method can improve real-time performance of the stereo matching as well as reducing cost of day-to-day and military high-precision image identification.

The invention discloses an agricultural weeding device. The agricultural weeding device comprises a weeding machine body, a connecting mechanism and a control mechanism, wherein the control mechanismis connected with the weeding machine body through the connecting mechanism; and the weeding machine body comprises a weeding rack, a weeding device, a cutting device and a collecting device, the weeding rack is used for carrying the connecting mechanism and the control mechanism, the weeding device is mounted on the bottom of the weeding rack and is used for removing weeds, the cutting device ismounted on the bottom of the weeding rack, is spaced from the weeding device, and is used for cutting the weeds, and the collecting device is arranged on one side of the weeding rack, and is used forcollecting and treating the weeds. The device can reduce the stress resistance and renewability of the weeds while removing the weeds, and stops weed regrowth.

The invention discloses a deep learning network application distribution self-assembly instruction processor core, a processor, a circuit and a processing method. The processor core comprises four register interface modules, and a preparation module, a convolution operation module and a pooling operation module are sequentially arranged between every two register interface modules; the processor comprises an instruction statistical distribution module which is configured to count deep convolutional network instructions and distribute instruction streams; the instruction statistics and distribution module is connected with the processor cores through the instruction stack module. And the instruction stack module is configured to receive and store the instruction streams allocated by the instruction statistical allocation module, perform acceleration operation of multiple instruction streams according to the stored instruction streams, control the processor core to form neural convolutional network architectures with different depths, and perform calculation processing.

The invention discloses a production process for preparing high-purity aluminum oxide by utilizing melt crystallization, and the production process comprises the following steps: introducing frozen saline water into the bottom of an outer jacket of a melting crystallizer, enabling the frozen saline water to uniformly flow on the outer wall of the crystallizer in the form of a falling film; addinga rough aluminum oxide solution into the melting crystallizer, flowing down the aluminum oxide solution on the inner wall of the crystallizer in the form of the falling film, cooling step by step to crystallize, growing out a uniform crystal layer on a heat exchange surface step by step, enabling concentration of the crystal layer to be higher than that of a mixture solution, enriching impuritiesinto the mixture solution step by step; regulating the hot water temperature through a heat exchanger for performing partial melting operation on the crystal layer, enabling low-melting-point impurities in the crystal layer to partly melt and remove, thereby further purifying the crystal. According to the method disclosed by the invention, the process is simple, the equipment structure is compact,the falling film form is adopted, energy is reduced, cost and equipment investment are reduced; and a product produced by the melting crystallizer is high in purity.

The invention discloses a production technology for preparing high-purity sorbic acid by melt crystallization. The production technology comprises the following steps of firstly, sending freezing salt water into the bottom part of an outer jacket of a melt crystallization device, enabling the freezing salt water to uniformly flow along the outer wall of the crystallization device in a falling film way, feeding a crude sorbic acid solution into the melt crystallization device, enabling the sorbic acid solution to flow down along the inner wall of the crystallization device in the falling film way, gradually cooling and crystallizing, gradually growing a uniform crystal layer on the heat exchange surface, enabling the concentration of the crystal layer to be higher than the concentration of mixture solution, and enabling impurities to gradually enrich into the mixture solution; then, melting one part of crystal layer by adjusting the hot water temperature through a heat exchanger, and melting and removing the impurities with low melting point in the crystal layer, so as to further purify crystals. The production technology has the advantages that the method and technology are simple, the equipment structure is compact, the falling film type is adopted, the energy is decreased, and the cost and equipment investment are reduced at same time; the purity of a product produced by the melt crystallization device is high.

The invention discloses a production process for preparing high purity dehydroacetic acid by virtue of melt crystallization. The process comprises the following steps of: firstly introducing frozen saline water into the bottom of an outer jacket of a melting crystallizer, wherein the frozen saline water uniformly flows in a falling film manner on the outer wall of the crystallizer; then inputting a concentrated dehydroacetic acid solution into the melting crystallizer, wherein the dehydroacetic acid solution flows in a falling film manner on the inner wall of the crystallizer and is gradually cooled to be crystallized; gradually growing a uniform crystal layer on a heat exchange face, wherein the concentration of the crystal layer is higher than that of a mixed solution, and impurities are gradually enriched in the mixture solution; and then adjusting the temperature of hot water by virtue of a heat exchanger to partly melt the crystal layer, namely a sweating operation, wherein low-melting-point impurities mingled in the crystal layer are partly melted and removed, so that the crystal is further purified. The process disclosed by the invention is simple, the equipment is compact in structure, and the energy is reduced in form of the falling film, and meanwhile, the cost and equipment investment are lowered; a product produced by the melting crystallizer is high in purity.

The invention provides a hip fracture fixing steel plate device with an adjustable supporting structure and a triangular mechanical stable form, and belongs to the technical field of hip fracture fixing. Comprising a steel plate, a guide part, an adjusting part, a limiting part and an anti-loosening part which are connected are arranged on the steel plate, and a supporting hole is formed in one side of the steel plate. The universality of a steel plate fixing model is improved, the difference of the fixing strength of a nail plate system caused by the bone anatomy difference of a patient is avoided, the clinical treatment level of the fracture patient is effectively guaranteed, the installation space of the supporting nails in the hip is effectively increased according to the hip structure, the overall stability effect of the steel plate is improved, and the stability of the whole steel plate is improved. The hip supporting strength of the steel plate fixing model is improved, the strength of the steel plate fixing model is further improved, adverse effects on a patient caused by fixing failure are avoided, and the use risk and the operation frequency of the patient are reduced.

The invention provides a circuit structure and method for expanding the number of operation times of a memory. The circuit structure comprises an N-bit cycle counter circuit, N address generation circuits and a memory array; the N address generation circuits are connected with the N-bit cycle counter circuit; the memory array is connected with the N address generation circuits; N times of operations on a memory word are dispersed into N different memory words; therefore, excessive operation on a memory word is avoided; when the power is off every time, the state of a trigger circuit is storedin a non-volatile phase change memory; when the power is on every time, the data in the non-volatile phase change memory is read out; therefore, the trigger circuit and the N-bit cycle counter circuitrecover to the state before the power is off; and thus, balanced operation of the memory words in any condition can be realized. By means of the circuit structure and method for expanding the numberof operation times of the memory provided by the invention, the problem that a phase change memory cannot realize the reading, writing and erasing balance of each memory unit in the prior art can be solved.

In a rewritable optical disk, effective use of a PCA area is inadequate, although effective use of the area is necessary, since securable area as a PCA area for getting the optimal recording power is limited and as a result, rewritable frequency is not sufficient. The distance W between last time used areas of a PCA area 311 of a first layer and a PCA area 321 of a second layer are calculated as above, and compared with a set threshold Wth. As a result of comparison, when the computed distance W is equal to threshold Wth as (C), or when the computed distance W is determined to be smaller than the distance Wth, DC erasure of PCA areas 311 and 321 are performed at that time. Thereby in terms of each of PCA areas 311 and 321, the small areas used for test writing are not inclined toward specific small areas and can be used uniformly. Therefore, the frequency of OPC operation can be increased until the degradation of PCA areas 311 and 321.

The invention discloses a load balancing method for prolonging the service life of a hot spot region, and relates to responding to an access request of a user to a hot spot region in a FLASH memory. The method comprises: according to a load balancing section global variable of current operation, operating the corresponding load balancing section, judging effectiveness of the current load balancingsection and correctness of read-write operation before the operation, completing read-write operation of the hot spot region, and updating the global variable and management information of the load balancing section of the current operation at the same time to prepare for the next read-write operation of the load balancing section. According to the method introduced in the invention, a pure software method can be used, the number of load balancing sections of the FLASH memory is configured, the read-write operation load of the hot spot region of the FLASH memory is dynamically balanced, the service life of the hot spot region is prolonged, the overall reliability of the FLASH is improved, the requirements of different users for the reliability of the FLASH memory, especially the bottleneck of the service life of the hot spot region are met, and the development efficiency is improved.

The invention provides a ferroelectric memory. The memory comprises: a first electrode layer, wherein the main crystallization direction of the first electrode layer comprises(110) or (220); a second electrode layer arranged opposite to the first electrode layer, wherein the main crystallization direction of the second electrode layer comprises (110) or (220); and a ferroelectric layer arranged between the first electrode layer and the second electrode layer, wherein the main crystallization direction of the ferroelectric layer is (111).