rice sorting In rice processing, from harvesting and drying to transfer and storage, raw materials are highly susceptible to mixing with various fine impurities. Among them, small stones and transparent glass fragments — nearly identical in size to ordinary rice grains — are the most troublesome impurities for rice mills. These impurities are tiny and closely resemble rice grains. They may seem insignificant, but they are a core pain point for many rice mills: ✅ Fine transparent glass fragments and small stones of the same particle size hide perfectly among rice grains. Ordinary color sorters have large recognition blind spots and suffer from severe missed detection.✅ Manual picking is time-consuming, labor-intensive, and extremely inefficient. Fine impurities are difficult to distinguish accurately by eye, and human fatigue easily leads to oversights.✅ Residual fine impurities are hard to detect, directly lowering the grade of finished rice while creating serious food safety hazards and damaging product reputation. In the traditional grain and oil sorting industry, it is widely believed that thoroughly removing fine stones and transparent glass fragments similar in size to rice grains can only be achieved with infrared color sorters. However, the high procurement cost of infrared equipment discourages many small and medium-sized rice mills. Is there a solution with comparable performance and higher cost-effectiveness? In this test, we conduct a real on-site machine trial to verify whether an AI HD Color Sorter can replace infrared equipment by accurately identifying and removing fine stones and mini glass fragments of the same size as rice grains. I. Test Materials: Rice + Small Stones & Glass FragmentsTo closely simulate actual factory production conditions, we manually and precisely prepared the test material, mixing two stubborn impurities into premium rice: fine sand and stones with particle sizes basically identical to rice grains, and mini transparent glass fragments. 【Caption: Core impurities in this test: small stones, transparent glass fragments】   II. Test Equipment: Next-Generation AI HD Color SorterThe machine tested is an AI HD Deep Learning Color Sorter, equipped with HD industrial lenses and deep learning AI algorithm model, breaking beyond the recognition limitations of traditional photoelectric sorting.With high-resolution imaging, even micron-level fine foreign matter is captured in complete detail. It accurately distinguishes normal rice grains from difficult-to-detect impurities such as transparent glass fragments and fine small stones. 【Caption: HD lenses imaging — every tiny detail revealed】   III. On-Machine Test Process: AI HD Precision SortingThe rice material mixed with stones and glass fragments is evenly fed into the AI HD Color Sorter’s inlet, and the machine performs fully automatic high-speed sorting.The HD lens captures the details of every particle in milliseconds, while the AI algorithm compares and analyzes in real time:✅ Normal premium rice → flows out through the finished product channel✅ Fine stones, transparent glass fragments → precisely rejected by air jets into the waste areaThe entire process runs automatically without manual intervention, with fast sorting speed and high precision. 【Caption: Sorting effect comparison: accept on the left, reject on the right】   IV. Test Results Comparison: Impurities Completely EliminatedAfter sorting, the results are clear at a glance:The small stones and transparent glass fragments mixed in the raw material are removed. The finished rice is clean and translucent, with no residual impurities, fully meeting the quality inspection standards for high-end rice.The test fully proves that in rice stone removal and glass fragment removal scenarios, the AI HD Color Sorter’s performance fully matches infrared equipment. 【Caption: Accept rice after sorting】   V. Core Advantages: AI HD Color Sorter — A Perfect Replacement for InfraredFor the most critical rice processing scenarios — removing stones and glass fragments — the AI HD Color Sorter, relying on HD industrial lenses to capture fine material details, offers strong replacement value: 1. Equivalent performance, maximum precision — HD industrial lenses capture the texture and contours of rice-grain-sized foreign matter, combined with AI deep learning algorithms, delivering sorting precision comparable to near-infrared equipment. 2. Lower price, significant cost reduction — No need to bear the high procurement cost of infrared equipment, enabling faster return on investment and profitability. 3. One machine, multiple uses, comprehensive quality improvement — In addition to stones and glass fragments, it can also remove moldy rice, yellow kernels, chalky kernels, and off-color rice grains.   【Caption: AI HD Color Sorter field test — rice sorting site】 The arrival of the AI HD Deep Learning Color Sorter breaks the industry’s fixed mindset that “only infrared can handle fine impurities in rice.”With equivalent sorting performance, lower procurement costs, and simpler operation and maintenance, every rice processing plant can achieve intelligent precision sorting at low cost.HAWIT AI HD Color Sorter — making every grain of rice stand up to inspection. Materials from different origins and categories have different impurity conditions.We welcome you to send samples for testing. We will provide one-on-one live machine demonstrations and free customized sorting solutions!

Rapeseed grains are extremely small, with diameters only in the millimeter range. The sprout defects from germination are even smaller in size, and their color is very close to that of sound rapeseeds, making them almost indistinguishable to the naked eye when mixed together.     The HAWIT AI HD Color Sorter is purpose-built for high-precision sorting of fine materials. 2026.09.07 High-Definition Imaging       High-Definition Imaging – Even the Smallest Sprouts Captured ClearlyEquipped with a high-definition industrial lens, the machine achieves a minimum resolution of 0.015 mm. The subtle morphological differences between sound rapeseeds and germinated sprouts are clearly visible under the lens – sprout tips, exposed embryos, and slight color variations are all captured with exceptional clarity. No matter how tiny the sprout, it is seen in perfect detail. AI       AI-Powered Intelligent Recognition – Not a Single Sprout MissedThe built-in deep learning algorithm automatically trains the sorting model for germinated seeds. Without the need for manual parameter adjustments, the AI automatically distinguishes between sound rapeseeds and sprouted ones. The more it sorts, the more accurate it becomes. Sorting Results       Rapeseed raw material is fed into the AI HD Color Sorter. The vibratory feeder evenly distributes the material as it enters the sorting channel. The HD lens scans each grain individually, and the AI makes millisecond-level decisions, precisely rejecting sprouted seeds. · Finished product: Sound rapeseeds – plump, free from germinated grains and impurities. · Rejects: Sprouted rapeseed grains – completely removed, not a single one left.     HAWIT AI HD Color Sorter – HD lenses see clearly, AI algorithms distinguish accurately. No matter how small the sprouts, they are sorted cleanly. In addition to rapeseed, the machine is equally suitable for sorting shape defects in various tiny seeds and minor grains. We welcome you to send samples for testing. HAWIT Technology – Making Sorting Smarter.

In garlic deep-processing production, the presence of skin-on garlic and spoiled garlic mixed in with the finished product are two major quality challenges. · Skin-on garlic: Cloves that are not completely peeled, with residual skin remaining on the surface. · Spoiled garlic: Cloves affected by mildew, black spots, or rot, posing food safety risks.   The HAWIT Belt Color Sorter is specifically designed for the characteristics of garlic materials, tailored for fragile, easily damaged peeled cloves, providing a stable and highly efficient fully automated sorting solution. In this real test, we thoroughly mixed finished peeled garlic cloves, skin-on garlic, and spoiled garlic to simulate the complex incoming material conditions of an actual factory, and fed them into the belt color sorter for sorting. accept       • Belt Conveying – Protecting Garlic IntegrityThe machine uses a flexible belt conveying structure, ensuring smooth material transport throughout the process. The material does not tumble or collide, significantly reducing the risk of secondary damage to garlic cloves. It maximizes the protection of intact cloves and effectively minimizes material loss. skin-on       • Intelligent Recognition – Precisely Removing Defective OnesRelying on an intelligent recognition system, the equipment accurately identifies defective items such as residual skin, surface browning, yellowing due to oxidation, breakage, and spoilage (blackening). It automatically separates good from bad materials, delivering stable and reliable sorting performance. spoiled       • Sorting ResultsThe finished peeled garlic cloves are plump, uniform in color, clean, and free of impurities. Skin-on garlic and spoiled cloves are precisely and effectively rejected. Sorting purity remains stable and controllable, thoroughly solving the industry pain points of manual sorting – inconsistency, high error rates, and low efficiency.   The machine can be directly integrated with garlic peeling production lines to enable a seamless automated workflow. It also supports flexible adjustment of sorting sensitivity and accuracy based on incoming material quality, adapting to various clove sizes and different processing grade requirements – significantly improving overall processing efficiency and finished product acceptance rates. For customized sorting solutions or on-site material testing, feel free to contact us anytime!☎  

In wheat processing, impurities come in many forms – transparent crushed glass, off-color wheat kernels, black and white stones, long and short straws, and wheat husks are all stubborn contaminants that are difficult to remove completely. They directly affect final product quality and market value.   In this test, we used a uniform set of test materials to conduct a side-by-side comparison of three color sorters with different technological approaches, giving you a clear look at their sorting capability.   Impurity List for This Test   2026 / 08 / 15 Impurity Type Characteristics Transparent crushed glass Colorless and transparent, different material from wheat Black stones Deep black, highly contrasting with wheat White stones Whitish, different shape and material from wheat Black rice Deep black, similar shape to wheat Brown wheat Darker color, same shape and material as wheat Long straws Elongated, yellowish in color Short straws Short straw pieces, different shape and material from wheat Wheat husks Similar size and color to wheat, different material   Test Rules   2026 / 08 / 15   RGB standard model & NIR model: Equal quantities of impurity samples were added. AI HD intelligent model: More impurities were added overall – with transparent crushed glass quantity doubled compared to the other two machines – to increase sorting difficulty. Hardware differences: RGB and NIR models are equipped with SD cameras; the AI model features an HD recognition camera.   Technology Principle & Test Results   2026 / 08 / 15 RGB       RGB Standard Color Sorter (SD camera) Relies on color and shape features to identify impurities. In this test, it successfully removed all impurities, meeting basic production standards. However, due to limited imaging resolution of the SD camera, for impurities with similar appearance to wheat (such as short straws and brown wheat), the judgment threshold had to be widened to ensure removal, resulting in a slightly higher carryover ratio. NIR       NIR Near-Infrared Color Sorter (SD camera + NIR) Adds near-infrared technology to RGB, enabling not only surface color impurity sorting but also identification of materials with internal defects. NIR adds a material property dimension, allowing more accurate recognition of "same-color, different-material" impurities without needing to expand the color threshold – achieving a lower carryover ratio. AI       AI HD Intelligent Color Sorter (HD camera + AI algorithms) As a next-generation high-end model, it is equipped with an HD imaging camera. In this test, we increased the difficulty by doubling the amount of glass impurities – yet it still removed all impurities with precision, demonstrating strong anti-interference capability. The HD camera provides clearer images, and AI algorithms deliver more accurate judgments, with a clear boundary between impurities and acceptable products – making it suitable for most production lines.     All three models deliver complete impurity removal with clean, uniform wheat output. The key differences are in carryover ratio and hardware — pick the one that fits your needs. Welcome for free sample testing — let us customize your ideal sorting solution! ✨

In the deep processing of crushed walnuts and pistachios, the presence of fine shell fragments is a common industry headache. After crushing, nut kernel fragments and tiny shell pieces are thoroughly mixed together – the particles are extremely small, making sorting exceptionally difficult:  Pistachio kernel fragments and shell fragments differ significantly in color, but the particles are so tiny that ordinary color sorters lack sufficient resolution – they can't see clearly, miss many particles, and easily leave shells behind.  Walnut kernel fragments and shell fragments are very close in color. Conventional equipment that relies solely on color detection simply cannot distinguish between kernel and shell, resulting in large amounts of shell contamination that severely affects final product quality. Traditional color sorters either have low pixel resolution, making fine particles hard to see, or rely only on color and fail to separate similar materials. Manual re-inspection is costly, material loss is high, and it's difficult to meet quality standards. Targeting the specific challenges of nut kernel processing, HAWIT's AI HD Color Sorter delivers a breakthrough solution specifically designed for separating fine shell fragments from kernel fragments.  Ultra-High-Definition Hardware – See Even the Smallest Impurities ClearlyThe machine is equipped with professional-grade HD industrial lenses, achieving a minimum resolution of 0.015 mm.This completely eliminates the common problem of "not being able to see fine particles" – no matter how fine the kernel pieces or how tiny the shell fragments, they are captured in high definition with complete detail.  AI-Powered Intelligent Recognition – Distinguishing "Material Essence," Not Just ColorWalnut kernel fragments and shell fragments are extremely similar in color, rendering ordinary color sorters completely ineffective.HAWIT's AI HD Color Sorter, however, does not rely solely on color judgment. Through deep learning of material texture features, it:Accurately captures the subtle difference between the rough texture of shell fragments and the smooth cut surfaces of kernel fragments – distinguishing kernel from shell by their essential characteristics. Sorting Results Pistachio fragments: Precisely separates fine kernel pieces from shell fragments – the finished product is pure pistachio kernel fragments, free from shells and impurities; all shell fragments are uniformly rejected.     Walnut fragments: Thoroughly solves the problem of sorting materials with similar colors – the finished product is highly pure walnut kernel fragments; all shell fragments are completely removed.   The core challenge in nut kernel processing has never been coarse material sorting, but the precise removal of fine particles, similarly colored materials, and trace impurities. HAWIT AI HD Color Sorter– Clear enough to see even the smallest particles, capable enough to separate even the closest colors –Truly achieving complete shell-kernel separation for nut kernels.Enabling deep-processed broken nuts to achieve high purity and consistently meet quality standards. For equipment sample videos, technical specifications, or free material testing, feel free to contact us anytime!

In pistachio processing, cracked shells are the most troublesome defect. Shells that are split, damaged, or even empty – these defective nuts are mixed in with sound kernels. Traditional color sorters use flat lenses that capture images only from the front and back angles. As pistachios tumble through the sorting channel, empty shells and cracked nuts hiding on the side are easily missed. To achieve acceptable results, the material must be run through the machine multiple times – consuming time and product, cutting throughput in half, and driving up processing costs.       That's why HAWIT proudly launches the brand-new AI HD Multi-View Sorter. Two sets of cross-mounted lenses, synchronized multi-camera imaging – truly achieving all-angle, full-coverage capture. One pass now delivers the same sorting result that used to take two passes on conventional equipment – upgrading both efficiency and accuracy.     • Core Hardware Upgrade – Say Goodbye to Blind Spots√ Upgraded to dual cross-mounted HD lenses√ Multi-camera, multi-angle synchronized imaging√ Three-dimensional capture of every material unit, 360° no-blind-spot coverage     • Key Advantages – One Pass, Multi-Angle Feature Capture, Efficiency DoubledConventional equipment: Defects hidden on the side → frequent misses → requires a second passHAWIT AI HD Multi-View Sorter: Single feed, multiple imaging passes, comprehensive screening   As pistachios tumble and move through the machine, their front, sides, back, and even recessed crevices are all precisely captured. No hidden defect can escape detection – fundamentally solving the problems of missed and misclassified rejects.     • Real Sorting Results √ Accept (170g): Plump, intact kernels with no cracks or empty shells – all meet export-grade standards. × Reject (35g): Effectively removed – broken split shells, shriveled empty nuts – with an extremely low carryover ratio and controlled material loss.     • Wide Adaptability – High-Performance Model for Multiple Nut Types The new AI HD Multi-View Sorter is not only optimized for pistachios but also perfectly suits various irregularly shaped, curved, or defect-prone nuts, such as: Almonds、Cashews、Pine nuts、Walnut kernels   • Tackles various hidden defects (cracked shells, empty shells, wormholes, etc.) • Deep removal of dark spots and color inconsistencies • 360° no-blind-spot sorting – eliminating missed rejects   One pass, double the results, zero blind spots. The top choice for nut processing plants looking to upgrade quality and throughput.   For sample testing, full sorting videos, detailed specifications, or custom solutions, feel free to DM us anytime! ✨  

At the dog days of summer, Hefei has entered "sauna mode". Yet at HAWIT's production workshop and shipping site, operations remain in full swing - another batch of color sorters has passed factory inspection and is officially on its way to customer sites!     🌡 No Matter How Hot It Gets, Quality Control Never "Down"As temperatures continue to soar during the dog days, HAWIT's uncompromising pursuit of quality has never slackened – not even by a fraction. Every machine must undergo rigorous testing before leaving the factory – with critical components such as the drive mechanism, imaging system, electrical systems, system functions, and software algorithms all thoroughly inspected one by one. After passing these checks, each unit must also clear 16 cosmetic inspection items to ensure that every machine arrives at the customer's production line in optimal condition.   💎 Technical Strength That Withstands the HeatThe July heat wave is sweeping across the country, but HAWIT's shipping pace has never slowed. No matter how hot it gets, we deliver on our promise!The reason we can keep shipping and delivering consistently through the dog days lies in our solid technical strength and quality assurance. 🔹 Proprietary R&D, Technology Leadership – The company focuses on the R&D and manufacturing of intelligent sorting equipment, with full in-house development across optics, mechanics, electronics, computing, and software. The supply chain is stable and reliable, unaffected by external environmental fluctuations. 🔹 AI Deep Learning Algorithms – The system automatically trains sorting models, achieving an accuracy rate of up to 99.9%. Equipped with HD cameras and high-pixel imaging, it delivers clearer images and maintains stable, precise recognition even with complex materials. 🔹 Comprehensive Systems, Global Service – After years of development, HAWIT has exported to over 60 countries and regions, established deep partnerships with numerous enterprises, and provides 24/7 remote operation and maintenance support.   The dog-day heat wave cannot stop HAWIT's shipping momentum, High temperatures won't lower our standards - quality builds our reputation. This summer, HAWIT ensures that every shipment withstands the "heat test"!

Why do traditional color sorters always require two passes when sorting wormhole-damaged chickpeas?   During the feeding process, materials inevitably collide and tumble. Traditional color sorters use flat lenses that capture images from only a single angle. As chickpeas tumble, the good side may be photographed while the defective side is missed - and wormhole-damaged beans are mistakenly judged as acceptable and end up mixed in with the finished product. To ensure accuracy, two sorting passes are often required - significantly reducing efficiency. HAWIT's new AI HD Multi-View Sorter features an upgraded design with cross lenses, completely solving this pain point. Cross-Mounted Lenses - Multi-Camera, Multi-Angle Imaging: The new machine replaces the original single set of flat lenses with two sets of cross-mounted lenses, enabling synchronized imaging from multiple cameras and multiple angles. As materials enter the sorting channel, the cross-mounted lenses capture images simultaneously - front, back, and sides. No matter where the wormhole is hidden, it can be accurately detected. One Pass Equals Two - Sorting Efficiency Doubled: Traditional flat-lens machines require two passes to achieve ideal accuracy when sorting chickpeas. The AI HD Multi-View Color Sorter completes the same level of precision in a single pass - multi-angle synchronized imaging combined with millisecond-level AI decision-making doubles sorting efficiency. Higher throughput in less time. Richer Data Acquisition, Faster Model Generation: Cross-mounted lenses capture richer and more complete image data, significantly accelerating AI algorithm labeling and speeding up the generation of new material models.   One pass delivers 360° no-blind-spot inspection. The job that used to take two passes now takes just one.   Flat lenses see one side, cross-mounted lenses see the whole picture. HAWIT AI HD Multi-View Color Sorter - one pass equals two, efficiency doubled.

Red kidney beans, plump and round, are an excellent ingredient for making porridge, fillings, and processed red bean products. But there is one defect that processing plants find most troublesome – wormholes in kidney beans. These wormholes are tiny, with a color nearly identical to that of sound beans. Traditional color sorters, which rely solely on color differentiation, are almost powerless against such "normally colored but defective" wormhole beans. The AI Deep Learning Color Sorter is designed specifically to tackle insect damage in beans. Sorting Made Easy in Four Steps – Wormhole Beans Have No Place to Hide Step 1: Feeding Red kidney beans are poured into the hopper. The vibratory feeder evenly spreads the material, allowing it to fall smoothly into the sorting channels. Step 2: Imaging High-definition lenses scan each bean one by one, with a minimum resolution of 0.015 mm. Wormholes and insect-bored holes are all captured. The system looks not only at color but also at shape and surface texture – this is the key to distinguishing "sound beans" from "wormhole beans." Step 3: Judgment The built-in deep learning algorithm automatically analyzes each bean without requiring manual parameter adjustments, completing the decision in milliseconds. The system continuously evolves during operation – the more it sorts, the more accurate it becomes. Step 4: Rejection The high-speed nozzle system responds precisely, instantly blowing wormhole beans into the reject channel. Traditional color sorters look at color; AI color sorters look at substance. Small as they are, wormholes make a big difference. The AI Deep Learning Color Sorter eliminates the "wormhole blind spot" in red kidney bean sorting – ensuring every finished bean is pure and reliable. HAWIT Technology – Making Sorting Smarter, Making Food Safer.

In rapeseed processing, off-color grains and fine impurities are key factors affecting final product quality. The 90-Channel Grain Color Sorter, with its core advantages of high throughput and high precision, provides an efficient solution for rapeseed sorting.     Ⅰ. High-Throughput Discharge for Rapeseed Addressing the characteristics of small grain size and large processing volumes, the 90-channel wide chute design ensures even material distribution, delivering large and stable discharge to meet the demands of continuous, large-scale production. Ⅱ. Intelligent Control for Rapeseed Sorting Equipped with a 15-inch high-resolution large screen, key parameters and throughput data for rapeseed sorting are displayed in real time, enabling easy operation and full visibility of production status. Ⅲ. High-Definiton Identification of Tiny Rapeseed Particles Using a CCD industrial lens combined with multi-spectral confocal technology, the minimum resolution reaches 0.015mm, clearly capturing fine off-color grains and impurities in rapeseed to ensure no omissions. Ⅳ. Precise Rejection of Impurities in Rapeseed The high-speed nozzle system responds quickly, accurately rejecting dark red, white, and other immature grains as well as various impurities. The finished product achieves high purity with an optimized carryover ratio, minimizing material loss.     The 90-Channel Grain Color Sorter, designed specifically for rapeseed sorting, helps processing companies improve finished product purity, reduce waste, and achieve efficient production with high throughput and high precision. HAWIT Technology --- Making Sorting More Intelligent

In the dog days of summer, the ground burns hot while the workshop buzzes with relentless energy.At the HAWIT production facility, packing and shipping never skip a beat —39 color sorters have completed debugging one after another, packed batch by batch, and are heading to customer sites overseas.     High temperatures won't stop our frontline — sweat builds quality.Production, assembly, debugging, securing, packing… every step is carried out with meticulous care. 39 machines, covering four main series:    Grain Color Sorters (21 units): 90-channel high-throughput design, specially engineered for small particles like rapeseed and millet — precisely removing off-color grains and impurities.  Rice Color Sorters (11 units): Tailored for rice processing, accurately eliminating off-color grains, chalky kernels, glass, plastic, and other hazardous impurities — safeguarding staple food safety.  AI Deep Learning Color Sorters (4 units): Equipped with high-definition lenses and self-training sorting models, tackling hidden defects such as wormholes, mold, off-color, and irregular shapes.  Mini Color Sorters (3 units): Compact size, high precision — ideal for small-batch processing or space-constrained environments, offering flexible deployment.     The scorching sun cannot melt our resolve; sweat cannot drown our passion.From components to complete machines, from debugging to packing — every step undergoes rigorous quality inspection. HAWIT color sorters have been exported to over 60 countries and regions worldwide, earning the trust of countless customers.   HAWIT Technology — making sorting smarter, and service warmer.  

In sunflower seed processing, wormholes and white spots are two of the most common types of hidden defects. Wormholes: Tiny holes with a color similar to that of normal kernels. White spots: Localized whitening on the seed surface, light in color and irregular in shape. These two defects directly lower the quality of the final product. The new AI Deep Learning Color Sorter, specially customized for sunflower seeds, accurately identifies wormholes and white spots. HAWIT’s Triple Engine – Pinpoints Every Defective Seed 1. Optical SensingEquipped with high-definition lenses, every sunflower seed is captured at high speed, producing clear images with fine detail. Tiny wormholes and light-colored white spots are fully revealed under the lens, providing high-quality image data for subsequent algorithm recognition. 2. Algorithmic Decision-MakingBuilt on a self-training model based on deep learning, no manual parameter adjustment is required. It integrates multi-dimensional features such as color, shape, and texture, greatly reducing the false positive rate and becoming more accurate with use. 3. Execution CoordinationAfter AI completes recognition and judgment within milliseconds, instructions are instantly transmitted to high-speed solenoid valves for precise rejection. This coordination mechanism breaks through the speed bottleneck of traditional color sorters, maintaining high throughput while improving rejection accuracy. The new AI Deep Learning Color Sorter – with high-definition vision as its core, deep learning algorithms as its brain, and high-speed coordinated execution as its hand – specializes in overcoming hidden defects like wormholes and white spots, making sunflower seed sorting clearer, more accurate, and more reliable.